Do not create venv each iteration & use separate data iterator for each parallel worker (#1522)

* Test passed

* CI

* Cache exchange

* Refine backtest scripts

* Minor

* Rename backtest file

* Add async mode for potential use

* Slient backtest. Add .

.github/labeler.yml

@@ -0,0 +1,6 @@
+documentation:
+- 'docs/**/*'
+- '**/*.md'
+
+waiting for triage:
+- any: ['**/*', '!docs/**/*', '!**/*.md']

.github/workflows/labeler.yml

@@ -0,0 +1,14 @@
+name: "Add label automatically"
+on:
+- pull_request_target
+
+jobs:
+  triage:
+    permissions:
+      contents: read
+      pull-requests: write
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/labeler@v4
+      with:
+        repo-token: "${{ secrets.GITHUB_TOKEN }}"

.github/workflows/test_macos.yml

@@ -1,94 +0,0 @@
-# There are some issues (in the downloading data phase) on MacOS when running with other tests. So we split it into an individual config.
-name: Test MacOS
-
-on:
-  push:
-    branches: [ main ]
-  pull_request:
-    branches: [ main ]
-
-jobs:
-  build:
-
-    runs-on: ${{ matrix.os }}
-    strategy:
-      matrix:
-        os: [macos-11, macos-latest]
-        # not supporting 3.6 due to annotations is not supported https://stackoverflow.com/a/52890129
-        python-version: [3.7, 3.8]
-
-    steps:
-    - uses: actions/checkout@v2
-
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
-      with:
-        python-version: ${{ matrix.python-version }}
-
-    - name: Lint with Black
-      run: |
-        cd ..
-        python -m pip install pip --upgrade
-        python -m pip install wheel --upgrade
-        python -m pip install black
-        python -m black qlib -l 120 --check --diff
-    # Test Qlib installed with pip
-
-    - name: Check Qlib with flake8
-      run: |
-        pip install --upgrade pip
-        pip install flake8
-        flake8 --ignore=E501,F541,E266,E402,W503,E731,E203 --per-file-ignores="__init__.py:F401,F403" qlib
-
-    - name: Install Qlib with pip
-      run: |
-          python -m pip install numpy==1.19.5
-          python -m pip install pyqlib --ignore-installed ruamel.yaml numpy
-    - name: Make html with sphnix
-      run: |
-        pip install -U sphinx
-        pip install sphinx_rtd_theme readthedocs_sphinx_ext
-        pip install --exists-action=w --no-cache-dir -r docs/requirements.txt 
-        cd docs 
-        sphinx-build -b html . build
-        cd ..
-    - name: Install Lightgbm for MacOS
-      run: |
-        /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Microsoft/qlib/main/.github/brew_install.sh)"
-        HOMEBREW_NO_AUTO_UPDATE=1 brew install lightgbm
-        # FIX MacOS error: Segmentation fault
-        # reference: https://github.com/microsoft/LightGBM/issues/4229
-        wget https://raw.githubusercontent.com/Homebrew/homebrew-core/fb8323f2b170bd4ae97e1bac9bf3e2983af3fdb0/Formula/libomp.rb
-        brew unlink libomp
-        brew install libomp.rb
-    - name: Test data downloads
-      run: |
-        python scripts/get_data.py qlib_data --name qlib_data_simple --target_dir ~/.qlib/qlib_data/cn_data_simple --interval 1d --region cn
-        python -c "import os; userpath=os.path.expanduser('~'); os.rename(userpath + '/.qlib/qlib_data/cn_data_simple', userpath + '/.qlib/qlib_data/cn_data')"
-        azcopy copy https://qlibpublic.blob.core.windows.net/data /tmp/qlibpublic --recursive
-        mv /tmp/qlibpublic/data tests/.data
-    - name: Test workflow by config (install from pip)
-      run: |
-        python qlib/workflow/cli.py examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
-        python -m pip uninstall -y pyqlib
-    # Test Qlib installed from source
-    - name: Install Qlib from source
-      run: |
-        python -m pip install --upgrade cython
-        python -m pip install numpy jupyter jupyter_contrib_nbextensions
-        python -m pip install -U scipy scikit-learn # installing without this line will cause errors on GitHub Actions, while instsalling locally won't
-        python -m pip install gym tianshou torch
-        pip install -e .
-    - name: Install test dependencies
-      run: |
-        python -m pip install --upgrade pip
-        python -m pip install -U pyopenssl idna
-        python -m pip install black pytest
-    - name: Unit tests with Pytest
-      run: |
-        pip install -r scripts/data_collector/pit/requirements.txt
-        cd tests
-        python -m pytest . --durations=0
-    - name: Test workflow by config (install from source)
-      run: |
-          python qlib/workflow/cli.py examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml

.github/workflows/test_qlib_from_pip.yml

@@ -0,0 +1,57 @@
+name: Test qlib from pip
+
+on:
+  push:
+    branches: [ main ]
+  pull_request:
+    branches: [ main ]
+
+jobs:
+  build:
+    timeout-minutes: 120
+
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [windows-latest, ubuntu-20.04, ubuntu-22.04, macos-11, macos-latest]
+        # not supporting 3.6 due to annotations is not supported https://stackoverflow.com/a/52890129
+        python-version: [3.7, 3.8]
+
+    steps:
+    - name: Test qlib from pip
+      uses: actions/checkout@v2
+
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v2
+      with:
+        python-version: ${{ matrix.python-version }}
+
+    - name: Update pip to the latest version
+      run: |
+        python -m pip install --upgrade pip
+
+    - name: Qlib installation test
+      run: |
+        python -m pip install pyqlib
+        # Specify the numpy version because the numpy upgrade caused the CI test to fail, 
+        # and this line of code will be removed when the next version of qlib is released.
+        python -m pip install "numpy<1.23"
+
+    - name: Install Lightgbm for MacOS
+      if: ${{ matrix.os == 'macos-11' || matrix.os == 'macos-latest' }}
+      run: |
+        /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Microsoft/qlib/main/.github/brew_install.sh)"
+        HOMEBREW_NO_AUTO_UPDATE=1 brew install lightgbm
+        # FIX MacOS error: Segmentation fault
+        # reference: https://github.com/microsoft/LightGBM/issues/4229
+        wget https://raw.githubusercontent.com/Homebrew/homebrew-core/fb8323f2b170bd4ae97e1bac9bf3e2983af3fdb0/Formula/libomp.rb
+        brew unlink libomp
+        brew install libomp.rb
+
+    - name: Downloads dependencies data
+      run: |
+        python scripts/get_data.py qlib_data --name qlib_data_simple --target_dir ~/.qlib/qlib_data/cn_data --interval 1d --region cn
+
+    - name: Test workflow by config
+      run: |
+        qrun examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml

.github/workflows/test_qlib_from_source.yml

@@ -1,4 +1,4 @@
-name: Test
+name: Test qlib from source
 
 on:
   push:
@@ -8,42 +8,61 @@ on:
 
 jobs:
   build:
+    timeout-minutes: 180
+    # we may retry for 3 times for `Unit tests with Pytest`
 
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [windows-latest, ubuntu-18.04, ubuntu-20.04]
+        os: [windows-latest, ubuntu-20.04, ubuntu-22.04, macos-11, macos-latest]
         # not supporting 3.6 due to annotations is not supported https://stackoverflow.com/a/52890129
         python-version: [3.7, 3.8]
 
     steps:
-    - uses: actions/checkout@v2
+    - name: Test qlib from source
+      uses: actions/checkout@v2
 
     - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
 
+    - name: Update pip to the latest version
+      # pip release version 23.1 on Apr.15 2023, CI failed to run, Please refer to #1495 ofr detailed logs.
+      # The pip version has been temporarily fixed to 23.0.1
+      run: |
+        python -m pip install pip==23.0.1
+
+    - name: Installing pytorch for macos
+      if: ${{ matrix.os == 'macos-11' || matrix.os == 'macos-latest' }}
+      run: |
+        python -m pip install torch torchvision torchaudio
+
+    - name: Installing pytorch for ubuntu
+      if: ${{ matrix.os == 'ubuntu-20.04' || matrix.os == 'ubuntu-22.04' }}
+      run: |
+        python -m pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cpu
+
+    - name: Installing pytorch for windows
+      if: ${{ matrix.os == 'windows-latest' }}
+      run: |
+        python -m pip install torch torchvision torchaudio
+
+    - name: Set up Python tools
+      run: |
+        python -m pip install --upgrade cython
+        python -m pip install -e .[dev]
+
     - name: Lint with Black
       run: |
-        pip install --upgrade pip
-        pip install black wheel
-        black qlib -l 120 --check --diff
-
-    - name: Install Qlib with pip
-      run: |
-        pip install numpy==1.19.5 ruamel.yaml
-        pip install pyqlib --ignore-installed
+        black . -l 120 --check --diff
 
     - name: Make html with sphinx
       run: |
-        pip install -U sphinx
-        pip install sphinx_rtd_theme readthedocs_sphinx_ext
-        pip install --exists-action=w --no-cache-dir -r docs/requirements.txt 
         cd docs 
-        sphinx-build -b html . build
+        sphinx-build -W --keep-going -b html . _build
         cd ..
-        
+
     # Check Qlib with pylint
     # TODO: These problems we will solve in the future. Important among them are: W0221, W0223, W0237, E1102
       # C0103: invalid-name
@@ -67,12 +86,11 @@ jobs:
       # W1309: f-string-without-interpolation
       # E1102: not-callable
       # E1136: unsubscriptable-object
-    # References for parameters: https://github.com/PyCQA/pylint/issues/4577#issuecomment-1000245962 
+    # References for parameters: https://github.com/PyCQA/pylint/issues/4577#issuecomment-1000245962
+    # We use sys.setrecursionlimit(2000) to make the recursion depth larger to ensure that pylint works properly (the default recursion depth is 1000).
     - name: Check Qlib with pylint
       run: |
-        pip install --upgrade pip
-        pip install pylint
-        pylint --disable=C0104,C0114,C0115,C0116,C0301,C0302,C0411,C0413,C1802,R0201,R0401,R0801,R0902,R0903,R0911,R0912,R0913,R0914,R0915,R1720,W0105,W0123,W0201,W0511,W0613,W1113,W1514,E0401,E1121,C0103,C0209,R0402,R1705,R1710,R1725,R1735,W0102,W0212,W0221,W0223,W0231,W0237,W0612,W0621,W0622,W0703,W1309,E1102,E1136 --const-rgx='[a-z_][a-z0-9_]{2,30}$' qlib --init-hook "import astroid; astroid.context.InferenceContext.max_inferred = 500"
+        pylint --disable=C0104,C0114,C0115,C0116,C0301,C0302,C0411,C0413,C1802,R0401,R0801,R0902,R0903,R0911,R0912,R0913,R0914,R0915,R1720,W0105,W0123,W0201,W0511,W0613,W1113,W1514,E0401,E1121,C0103,C0209,R0402,R1705,R1710,R1725,R1735,W0102,W0212,W0221,W0223,W0231,W0237,W0612,W0621,W0622,W0703,W1309,E1102,E1136 --const-rgx='[a-z_][a-z0-9_]{2,30}$' qlib --init-hook "import astroid; astroid.context.InferenceContext.max_inferred = 500; import sys; sys.setrecursionlimit(2000)"
 
     # The following flake8 error codes were ignored:
       # E501 line too long
@@ -95,47 +113,52 @@ jobs:
         # Description: If there is whitespace before ":", it cannot pass the black check.
     - name: Check Qlib with flake8
       run: |
-        pip install --upgrade pip
-        pip install flake8
         flake8 --ignore=E501,F541,E266,E402,W503,E731,E203 --per-file-ignores="__init__.py:F401,F403" qlib
 
     # https://github.com/python/mypy/issues/10600
     - name: Check Qlib with mypy
       run: |
-        pip install mypy
         mypy qlib --install-types --non-interactive || true
-        mypy qlib
+        mypy qlib --verbose
+    
+    - name: Check Qlib ipynb with nbqa
+      run: |
+        nbqa black . -l 120 --check --diff
+        nbqa pylint . --disable=C0104,C0114,C0115,C0116,C0301,C0302,C0411,C0413,C1802,R0401,R0801,R0902,R0903,R0911,R0912,R0913,R0914,R0915,R1720,W0105,W0123,W0201,W0511,W0613,W1113,W1514,E0401,E1121,C0103,C0209,R0402,R1705,R1710,R1725,R1735,W0102,W0212,W0221,W0223,W0231,W0237,W0612,W0621,W0622,W0703,W1309,E1102,E1136,W0719,W0104,W0404,C0412,W0611,C0410 --const-rgx='[a-z_][a-z0-9_]{2,30}$'
 
     - name: Test data downloads
       run: |
-        python scripts/get_data.py qlib_data --name qlib_data_simple --target_dir ~/.qlib/qlib_data/cn_data_simple --interval 1d --region cn
-        python -c "import os; userpath=os.path.expanduser('~'); os.rename(userpath + '/.qlib/qlib_data/cn_data_simple', userpath + '/.qlib/qlib_data/cn_data')"
-        azcopy copy https://qlibpublic.blob.core.windows.net/data /tmp/qlibpublic --recursive
+        python scripts/get_data.py qlib_data --name qlib_data_simple --target_dir ~/.qlib/qlib_data/cn_data --interval 1d --region cn
+        azcopy copy https://qlibpublic.blob.core.windows.net/data/rl /tmp/qlibpublic/data --recursive
         mv /tmp/qlibpublic/data tests/.data
 
-    - name: Test workflow by config (install from pip)
+    - name: Install Lightgbm for MacOS
+      if: ${{ matrix.os == 'macos-11' || matrix.os == 'macos-latest' }}
       run: |
-        python qlib/workflow/cli.py examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
-        python -m pip uninstall -y pyqlib
+        /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Microsoft/qlib/main/.github/brew_install.sh)"
+        HOMEBREW_NO_AUTO_UPDATE=1 brew install lightgbm
+        # FIX MacOS error: Segmentation fault
+        # reference: https://github.com/microsoft/LightGBM/issues/4229
+        wget https://raw.githubusercontent.com/Homebrew/homebrew-core/fb8323f2b170bd4ae97e1bac9bf3e2983af3fdb0/Formula/libomp.rb
+        brew unlink libomp
+        brew install libomp.rb
 
-     # Test Qlib installed from source
-    - name: Install Qlib from source
+    # Run after data downloads
+    - name: Check Qlib ipynb with nbconvert
       run: |
-        pip install --upgrade cython jupyter jupyter_contrib_nbextensions numpy scipy scikit-learn # installing without this line will cause errors on GitHub Actions, while instsalling locally won't
-        pip install gym tianshou torch
-        pip install -e .
-
-    - name: Install test dependencies
-      run: |
-        pip install --upgrade pip
-        pip install black pytest
-
-    - name: Unit tests with Pytest
-      run: |
-        pip install -r scripts/data_collector/pit/requirements.txt
-        cd tests
-        python -m pytest . --durations=10
+        # add more ipynb files in future
+        jupyter nbconvert --to notebook --execute examples/workflow_by_code.ipynb
 
     - name: Test workflow by config (install from source)
       run: |
+        python -m pip install numba
         python qlib/workflow/cli.py examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
+
+    - name: Unit tests with Pytest
+      uses: nick-fields/retry@v2
+      with:
+        timeout_minutes: 60
+        max_attempts: 3
+        command: |
+          cd tests
+          python -m pytest . -m "not slow" --durations=0

.github/workflows/test_qlib_from_source_slow.yml

@@ -0,0 +1,60 @@
+name: Test qlib from source slow
+
+on:
+  push:
+    branches: [ main ]
+  pull_request:
+    branches: [ main ]
+
+jobs:
+  build:
+    timeout-minutes: 720
+    # we may retry for 3 times for `Unit tests with Pytest`
+
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [windows-latest, ubuntu-20.04, ubuntu-22.04, macos-11, macos-latest]
+        # not supporting 3.6 due to annotations is not supported https://stackoverflow.com/a/52890129
+        python-version: [3.7, 3.8]
+
+    steps:
+    - name: Test qlib from source slow
+      uses: actions/checkout@v2
+
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v2
+      with:
+        python-version: ${{ matrix.python-version }}
+
+    - name: Set up Python tools
+      # pip release version 23.1 on Apr.15 2023, CI failed to run, Please refer to #1495 ofr detailed logs.
+      # The pip version has been temporarily fixed to 23.0.1
+      run: |
+        python -m pip install pip==23.0.1
+        pip install --upgrade cython numpy
+        pip install -e .[dev]
+
+    - name: Downloads dependencies data
+      run: |
+        python scripts/get_data.py qlib_data --name qlib_data_simple --target_dir ~/.qlib/qlib_data/cn_data --interval 1d --region cn
+
+    - name: Install Lightgbm for MacOS
+      if: ${{ matrix.os == 'macos-11' || matrix.os == 'macos-latest' }}
+      run: |
+        /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Microsoft/qlib/main/.github/brew_install.sh)"
+        HOMEBREW_NO_AUTO_UPDATE=1 brew install lightgbm
+        # FIX MacOS error: Segmentation fault
+        # reference: https://github.com/microsoft/LightGBM/issues/4229
+        wget https://raw.githubusercontent.com/Homebrew/homebrew-core/fb8323f2b170bd4ae97e1bac9bf3e2983af3fdb0/Formula/libomp.rb
+        brew unlink libomp
+        brew install libomp.rb
+
+    - name: Unit tests with Pytest
+      uses: nick-fields/retry@v2
+      with:
+        timeout_minutes: 240
+        max_attempts: 3
+        command: |
+          cd tests
+          python -m pytest . -m "slow" --durations=0

.gitignore

@@ -10,7 +10,6 @@ _build
 build/
 dist/
 
-
 *.pkl
 *.hd5
 *.csv
@@ -24,6 +23,11 @@ qlib/VERSION.txt
 qlib/data/_libs/expanding.cpp
 qlib/data/_libs/rolling.cpp
 examples/estimator/estimator_example/
+examples/rl/data/
+examples/rl/checkpoints/
+examples/rl/outputs/
+examples/rl_order_execution/data/
+examples/rl_order_execution/outputs/
 
 *.egg-info/
 
@@ -44,3 +48,4 @@ tags
 *.swp
 
 ./pretrain
+.idea/

.mypy.ini

@@ -1,6 +1,6 @@
 [mypy]
 exclude = (?x)(
-    ^qlib/backtest
+    ^qlib/backtest/high_performance_ds\.py$
     | ^qlib/contrib
     | ^qlib/data
     | ^qlib/model

.pre-commit-config.yaml

@@ -1,6 +1,6 @@
 repos:
 -   repo: https://github.com/psf/black
-    rev: 22.1.0
+    rev: 22.6.0
     hooks:
     -   id: black
         args: ["qlib", "-l 120"]

CHANGES.rst

@@ -1,63 +1,63 @@
 Changelog
-====================
+=========
 Here you can see the full list of changes between each QLib release.
 
 Version 0.1.0
---------------------
+-------------
 This is the initial release of QLib library.
 
 Version 0.1.1
---------------------
+-------------
 Performance optimize. Add more features and operators.
 
 Version 0.1.2
---------------------
-- Support operator syntax. Now ``High() - Low()`` is equivalent to ``Sub(High(), Low())``.   
+-------------
+- Support operator syntax. Now ``High() - Low()`` is equivalent to ``Sub(High(), Low())``.
 - Add more technical indicators.
 
 Version 0.1.3
---------------------
+-------------
 Bug fix and add instruments filtering mechanism.
 
 Version 0.2.0
---------------------
+-------------
 - Redesign ``LocalProvider`` database format for performance improvement.
 - Support load features as string fields.
 - Add scripts for database construction.
 - More operators and technical indicators.
 
 Version 0.2.1
---------------------
+-------------
 - Support registering user-defined ``Provider``.
 - Support use operators in string format, e.g. ``['Ref($close, 1)']`` is valid field format.
 - Support dynamic fields in ``$some_field`` format. And existing fields like ``Close()`` may be deprecated in the future.
 
 Version 0.2.2
---------------------
+-------------
 - Add ``disk_cache`` for reusing features (enabled by default).
 - Add ``qlib.contrib`` for experimental model construction and evaluation.
 
 
 Version 0.2.3
---------------------
+-------------
 - Add ``backtest`` module
 - Decoupling the Strategy, Account, Position, Exchange from the backtest module
 
 Version 0.2.4
---------------------
+-------------
 - Add ``profit attribution`` module
 - Add ``rick_control`` and ``cost_control`` strategies
-  
+
 Version 0.3.0
---------------------
+-------------
 - Add ``estimator`` module
 
 Version 0.3.1
---------------------
+-------------
 - Add ``filter`` module
 
 Version 0.3.2
---------------------
+-------------
 - Add real price trading, if the ``factor`` field in the data set is incomplete, use ``adj_price`` trading
 - Refactor ``handler`` ``launcher`` ``trainer`` code
 - Support ``backtest`` configuration parameters in the configuration file
@@ -65,16 +65,16 @@ Version 0.3.2
 - Fix bug of ``filter`` module
 
 Version 0.3.3
--------------------
+-------------
 - Fix bug of ``filter`` module
 
 Version 0.3.4
---------------------
+-------------
 - Support for ``finetune model``
 - Refactor ``fetcher`` code
 
 Version 0.3.5
---------------------
+-------------
 - Support multi-label training, you can provide multiple label in ``handler``. (But LightGBM doesn't support due to the algorithm itself)
 - Refactor ``handler`` code, dataset.py is no longer used, and you can deploy your own labels and features in ``feature_label_config``
 - Handler only offer DataFrame. Also, ``trainer`` and model.py only receive DataFrame
@@ -82,10 +82,10 @@ Version 0.3.5
 - Move some date config from ``handler`` to ``trainer``
 
 Version 0.4.0
---------------------
+-------------
 - Add `data` package that holds all data-related codes
 - Reform the data provider structure
-- Create a server for data centralized management `qlib-server<https://amc-msra.visualstudio.com/trading-algo/_git/qlib-server>`_
+- Create a server for data centralized management `qlib-server <https://amc-msra.visualstudio.com/trading-algo/_git/qlib-server>`_
 - Add a `ClientProvider` to work with server
 - Add a pluggable cache mechanism
 - Add a recursive backtracking algorithm to inspect the furthest reference date for an expression
@@ -100,7 +100,7 @@ Version 0.4.0
 
 
 Version 0.4.1
---------------------
+-------------
 - Add support Windows
 - Fix ``instruments`` type bug
 - Fix ``features`` is empty bug(It will cause failure in updating)
@@ -112,19 +112,19 @@ Version 0.4.1
 
 
 Version 0.4.2
---------------------
+-------------
 - Refactor DataHandler
 - Add ``Alpha360`` DataHandler
 
 
 Version 0.4.3
---------------------
+-------------
 - Implementing Online Inference and Trading Framework
 - Refactoring The interfaces of backtest and strategy module.
 
 
 Version 0.4.4
---------------------
+-------------
 - Optimize cache generation performance
 - Add report module
 - Fix bug when using ``ServerDatasetCache`` offline.
@@ -138,7 +138,7 @@ Version 0.4.4
 
 
 Version 0.4.5
---------------------
+-------------
 - Add multi-kernel implementation for both client and server.
     - Support a new way to load data from client which skips dataset cache.
     - Change the default dataset method from single kernel implementation to multi kernel implementation.
@@ -146,14 +146,14 @@ Version 0.4.5
 - Support a new method to write config file by using dict.
 
 Version 0.4.6
---------------------
+-------------
 - Some bugs are fixed
     - The default config in `Version 0.4.5` is not friendly to daily frequency data.
     - Backtest error in TopkWeightStrategy when `WithInteract=True`.
 
 
 Version 0.5.0
---------------------
+-------------
 - First opensource version
     - Refine the docs, code
     - Add baselines
@@ -161,19 +161,19 @@ Version 0.5.0
 
 
 Version 0.8.0
---------------------
+-------------
 - The backtest is greatly refactored.
     - Nested decision execution framework is supported
     - There are lots of changes for daily trading, it is hard to list all of them. But a few important changes could be noticed
         - The trading limitation is more accurate;
-            - In `previous version <https://github.com/microsoft/qlib/blob/v0.7.2/qlib/contrib/backtest/exchange.py#L160>`_, longing and shorting actions share the same action.
-            - In `current version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/backtest/exchange.py#L304>`_, the trading limitation is different between logging and shorting action.
+            - In `previous version <https://github.com/microsoft/qlib/blob/v0.7.2/qlib/contrib/backtest/exchange.py#L160>`__, longing and shorting actions share the same action.
+            - In `current version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/backtest/exchange.py#L304>`__, the trading limitation is different between logging and shorting action.
         - The constant is different when calculating annualized metrics.
-            - `Current version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/contrib/evaluate.py#L42>`_ uses more accurate constant than `previous version <https://github.com/microsoft/qlib/blob/v0.7.2/qlib/contrib/evaluate.py#L22>`_
-        - `A new version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/tests/data.py#L17>`_ of data is released. Due to the unstability of Yahoo data source, the data may be different after downloading data again.
-        - Users could check out the backtesting results between  `Current version <https://github.com/microsoft/qlib/tree/7c31012b507a3823117bddcc693fc64899460b2a/examples/benchmarks>`_ and `previous version <https://github.com/microsoft/qlib/tree/v0.7.2/examples/benchmarks>`_
+            - `Current version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/contrib/evaluate.py#L42>`_ uses more accurate constant than `previous version <https://github.com/microsoft/qlib/blob/v0.7.2/qlib/contrib/evaluate.py#L22>`__
+        - `A new version <https://github.com/microsoft/qlib/blob/7c31012b507a3823117bddcc693fc64899460b2a/qlib/tests/data.py#L17>`__ of data is released. Due to the unstability of Yahoo data source, the data may be different after downloading data again.
+        - Users could check out the backtesting results between  `Current version <https://github.com/microsoft/qlib/tree/7c31012b507a3823117bddcc693fc64899460b2a/examples/benchmarks>`__ and `previous version <https://github.com/microsoft/qlib/tree/v0.7.2/examples/benchmarks>`__
 
 
 Other Versions
-----------------------------------
+--------------
 Please refer to `Github release Notes <https://github.com/microsoft/qlib/releases>`_

README.md

@@ -11,6 +11,8 @@
 Recent released features
 | Feature | Status |
 | --                      | ------    |
+| Release Qlib v0.9.0 | :octocat: [Released](https://github.com/microsoft/qlib/releases/tag/v0.9.0) on Dec 9, 2022 |
+| RL Learning Framework | :hammer: :chart_with_upwards_trend: Released on Nov 10, 2022. [#1332](https://github.com/microsoft/qlib/pull/1332), [#1322](https://github.com/microsoft/qlib/pull/1322), [#1316](https://github.com/microsoft/qlib/pull/1316),[#1299](https://github.com/microsoft/qlib/pull/1299),[#1263](https://github.com/microsoft/qlib/pull/1263), [#1244](https://github.com/microsoft/qlib/pull/1244), [#1169](https://github.com/microsoft/qlib/pull/1169), [#1125](https://github.com/microsoft/qlib/pull/1125), [#1076](https://github.com/microsoft/qlib/pull/1076)|
 | HIST and IGMTF models | :chart_with_upwards_trend: [Released](https://github.com/microsoft/qlib/pull/1040) on Apr 10, 2022 |
 | Qlib [notebook tutorial](https://github.com/microsoft/qlib/tree/main/examples/tutorial) | 📖 [Released](https://github.com/microsoft/qlib/pull/1037) on Apr 7, 2022 | 
 | Ibovespa index data | :rice: [Released](https://github.com/microsoft/qlib/pull/990) on Apr 6, 2022 |
@@ -40,13 +42,11 @@ Features released before 2021 are not listed here.
   <img src="http://fintech.msra.cn/images_v070/logo/1.png" />
 </p>
 
+Qlib is an open-source, AI-oriented quantitative investment platform that aims to realize the potential, empower research, and create value using AI technologies in quantitative investment, from exploring ideas to implementing productions. Qlib supports diverse machine learning modeling paradigms, including supervised learning, market dynamics modeling, and reinforcement learning.
 
-Qlib is an AI-oriented quantitative investment platform, which aims to realize the potential, empower the research, and create the value of AI technologies in quantitative investment.
+An increasing number of SOTA Quant research works/papers in diverse paradigms are being released in Qlib to collaboratively solve key challenges in quantitative investment. For example, 1) using supervised learning to mine the market's complex non-linear patterns from rich and heterogeneous financial data, 2) modeling the dynamic nature of the financial market using adaptive concept drift technology, and 3) using reinforcement learning to model continuous investment decisions and assist investors in optimizing their trading strategies.
 
 It contains the full ML pipeline of data processing, model training, back-testing; and covers the entire chain of quantitative investment: alpha seeking, risk modeling, portfolio optimization, and order execution. 
-
-With Qlib, users can easily try ideas to create better Quant investment strategies.
-
 For more details, please refer to our paper ["Qlib: An AI-oriented Quantitative Investment Platform"](https://arxiv.org/abs/2009.11189).
 
 
@@ -67,6 +67,7 @@ For more details, please refer to our paper ["Qlib: An AI-oriented Quantitative
             <li type="circle"><a href="#auto-quant-research-workflow">Auto Quant Research Workflow</a></li>
             <li type="circle"><a href="#building-customized-quant-research-workflow-by-code">Building Customized Quant Research Workflow by Code</a></li></ul>
         <li><a href="#quant-dataset-zoo"><strong>Quant Dataset Zoo</strong></a></li>
+        <li><a href="#learning-framework">Learning Framework</a></li>
         <li><a href="#more-about-qlib">More About Qlib</a></li>
         <li><a href="#offline-mode-and-online-mode">Offline Mode and Online Mode</a>
         <ul>
@@ -105,21 +106,16 @@ Your feedbacks about the features are very important.
 # Framework of Qlib
 
 <div style="align: center">
-<img src="docs/_static/img/framework.svg" />
+<img src="docs/_static/img/framework-abstract.jpg" />
 </div>
 
-At the module level, Qlib is a platform that consists of the above components. The components are designed as loose-coupled modules, and each component could be used stand-alone.
+The high-level framework of Qlib can be found above(users can find the [detailed framework](https://qlib.readthedocs.io/en/latest/introduction/introduction.html#framework) of Qlib's design when getting into nitty gritty).
+The components are designed as loose-coupled modules, and each component could be used stand-alone.
 
-| Name                   | Description                                                                                                                                                                                                                                                                                                                                                             |
-| ------                 | -----                                                                                                                                                                                                                                                                                                                                                                   |
-| `Infrastructure` layer | `Infrastructure` layer provides underlying support for Quant research. `DataServer` provides a high-performance infrastructure for users to manage and retrieve raw data. `Trainer` provides a flexible interface to control the training process of models, which enable algorithms to control the training process.                                                       |
-| `Workflow` layer       | `Workflow` layer covers the whole workflow of quantitative investment. `Information Extractor` extracts data for models. `Forecast Model` focuses on producing all kinds of forecast signals (e.g. _alpha_, risk) for other modules. With these signals `Decision Generator` will generate the target trading decisions(i.e. portfolio, orders)  to be executed by `Execution Env` (i.e. the trading market).  There may be multiple levels of `Trading Agent` and `Execution Env` (e.g. an _order executor trading agent and intraday order execution environment_ could behave like an interday trading environment and nested in  _daily portfolio management trading agent and interday trading environment_  ) |
-| `Interface` layer      | `Interface` layer tries to present a user-friendly interface for the underlying system. `Analyser` module will provide users detailed analysis reports of forecasting signals, portfolios and execution results                                                                                                                                                                 |
-
-* The modules with hand-drawn style are under development and will be released in the future.
-* The modules with dashed borders are highly user-customizable and extendible.
-
-(p.s. framework image is created with https://draw.io/)
+Qlib provides a strong infrastructure to support Quant research. [Data](https://qlib.readthedocs.io/en/latest/component/data.html) is always an important part.
+A strong learning framework is designed to support diverse learning paradigms (e.g. [reinforcement learning](https://qlib.readthedocs.io/en/latest/component/rl.html), [supervised learning](https://qlib.readthedocs.io/en/latest/component/workflow.html#model-section)) and patterns at different levels(e.g. [market dynamic modeling](https://qlib.readthedocs.io/en/latest/component/meta.html)).
+By modeling the market, [trading strategies](https://qlib.readthedocs.io/en/latest/component/strategy.html) will generate trade decisions that will be executed. Multiple trading strategies and executors in different levels or granularities can be [nested to be optimized and run together](https://qlib.readthedocs.io/en/latest/component/highfreq.html).
+At last, a comprehensive [analysis](https://qlib.readthedocs.io/en/latest/component/report.html) will be provided and the model can be [served online](https://qlib.readthedocs.io/en/latest/component/online.html) in a low cost.
 
 
 # Quick Start
@@ -170,12 +166,25 @@ Also, users can install the latest dev version ``Qlib`` by the source code accor
     git clone https://github.com/microsoft/qlib.git && cd qlib
     pip install .
     ```
-  **Note**:  You can install Qlib with `python setup.py install` as well. But it is not the recommanded approach. It will skip `pip` and cause obscure problems. For example, **only** the command ``pip install .`` **can** overwrite the stable version installed by ``pip install pyqlib``, while the command ``python setup.py install`` **can't**.
+  **Note**:  You can install Qlib with `python setup.py install` as well. But it is not the recommended approach. It will skip `pip` and cause obscure problems. For example, **only** the command ``pip install .`` **can** overwrite the stable version installed by ``pip install pyqlib``, while the command ``python setup.py install`` **can't**.
 
-**Tips**: If you fail to install `Qlib` or run the examples in your environment,  comparing your steps and the [CI workflow](.github/workflows/test.yml) may help you find the problem.
+**Tips**: If you fail to install `Qlib` or run the examples in your environment,  comparing your steps and the [CI workflow](.github/workflows/test_qlib_from_source.yml) may help you find the problem.
 
 ## Data Preparation
 Load and prepare data by running the following code:
+
+### Get with module
+  ```bash
+  # get 1d data
+  python -m qlib.run.get_data qlib_data --target_dir ~/.qlib/qlib_data/cn_data --region cn
+
+  # get 1min data
+  python -m qlib.run.get_data qlib_data --target_dir ~/.qlib/qlib_data/cn_data_1min --region cn --interval 1min
+
+  ```
+
+### Get from source
+
   ```bash
   # get 1d data
   python scripts/get_data.py qlib_data --target_dir ~/.qlib/qlib_data/cn_data --region cn
@@ -197,6 +206,8 @@ We recommend users to prepare their own data if they have a high-quality dataset
   > 
   > It is recommended that users update the data manually once (--trading_date 2021-05-25) and then set it to update automatically.
   >
+  > **NOTE**: Users can't incrementally  update data based on the offline data provided by Qlib(some fields are removed to reduce the data size). Users should use [yahoo collector](https://github.com/microsoft/qlib/tree/main/scripts/data_collector/yahoo#automatic-update-of-daily-frequency-datafrom-yahoo-finance) to download Yahoo data from scratch and then incrementally update it.
+  > 
   > For more information, please refer to: [yahoo collector](https://github.com/microsoft/qlib/tree/main/scripts/data_collector/yahoo#automatic-update-of-daily-frequency-datafrom-yahoo-finance)
 
   * Automatic update of data to the "qlib" directory each trading day(Linux)
@@ -389,6 +400,17 @@ Dataset plays a very important role in Quant. Here is a list of the datasets bui
 [Here](https://qlib.readthedocs.io/en/latest/advanced/alpha.html) is a tutorial to build dataset with `Qlib`.
 Your PR to build new Quant dataset is highly welcomed.
 
+
+# Learning Framework
+Qlib is high customizable and a lot of its components are learnable.
+The learnable components are instances of `Forecast Model` and `Trading Agent`. They are learned based on the `Learning Framework` layer and then applied to multiple scenarios in `Workflow` layer.
+The learning framework leverages the `Workflow` layer as well(e.g. sharing `Information Extractor`, creating environments based on `Execution Env`).
+
+Based on learning paradigms, they can be categorized into reinforcement learning and supervised learning.
+- For supervised learning, the detailed docs can be found [here](https://qlib.readthedocs.io/en/latest/component/model.html).
+- For reinforcement learning, the detailed docs can be found [here](https://qlib.readthedocs.io/en/latest/component/rl.html). Qlib's RL learning framework leverages `Execution Env` in `Workflow` layer to create environments.  It's worth noting that `NestedExecutor` is supported as well. This empowers users to optimize different level of strategies/models/agents together (e.g. optimizing an order execution strategy for a specific portfolio management strategy).
+
+
 # More About Qlib
 If you want to have a quick glance at the most frequently used components of qlib, you can try notebooks [here](examples/tutorial/).
 
@@ -458,7 +480,7 @@ Before we released Qlib as an open-source project on Github in Sep 2020, Qlib is
 
 This project welcomes contributions and suggestions.  
 **Here are some 
-[code standards](docs/developer/code_standard.rst) for submiting a pull request.**
+[code standards and development guidance](docs/developer/code_standard_and_dev_guide.rst) for submiting a pull request.**
 
 Making contributions is not a hard thing. Solving an issue(maybe just answering a question raised in [issues list](https://github.com/microsoft/qlib/issues) or [gitter](https://gitter.im/Microsoft/qlib)), fixing/issuing a bug, improving the documents and even fixing a typo are important contributions to Qlib.
 

docs/FAQ/FAQ.rst

@@ -3,7 +3,7 @@ Qlib FAQ
 ############
 
 Qlib Frequently Asked Questions
-================================
+===============================
 .. contents::
     :depth: 1
     :local:
@@ -13,7 +13,7 @@ Qlib Frequently Asked Questions
 
 
 1. RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase...
-------------------------------------------------------------------------------------------------------------------------------------
+-----------------------------------------------------------------------------------------------------------------------------------
 
 .. code-block:: console
 
@@ -52,7 +52,7 @@ This is caused by the limitation of multiprocessing under windows OS. Please ref
 
 
 2. qlib.data.cache.QlibCacheException: It sees the key(...) of the redis lock has existed in your redis db now.
------------------------------------------------------------------------------------------------------------------
+---------------------------------------------------------------------------------------------------------------
 
 It sees the key of the redis lock has existed in your redis db now. You can use the following command to clear your redis keys and rerun your commands
 
@@ -72,7 +72,7 @@ If the issue is not resolved, use ``keys *`` to find if multiple keys exist. If
 Also, feel free to post a new issue in our GitHub repository. We always check each issue carefully and try our best to solve them.
 
 3. ModuleNotFoundError: No module named 'qlib.data._libs.rolling'
-------------------------------------------------------------------------------------------------------------------------------------
+-----------------------------------------------------------------
 
 .. code-block:: python
 
@@ -101,7 +101,7 @@ Also, feel free to post a new issue in our GitHub repository. We always check ea
 
 
 4. BadNamespaceError: / is not a connected namespace
-------------------------------------------------------------------------------------------------------------------------------------
+----------------------------------------------------
 
 .. code-block:: python
 
@@ -125,7 +125,7 @@ Also, feel free to post a new issue in our GitHub repository. We always check ea
 
 
 5. TypeError: send() got an unexpected keyword argument 'binary'
-------------------------------------------------------------------------------------------------------------------------------------
+----------------------------------------------------------------
 
 .. code-block:: python
 

docs/Makefile

@@ -17,4 +17,5 @@ help:
 # Catch-all target: route all unknown targets to Sphinx using the new
 # "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
 %: Makefile
+	pip install -r requirements.txt
 	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

docs/advanced/PIT.rst

@@ -1,14 +1,14 @@
 .. _pit:
 
-===========================
+============================
 (P)oint-(I)n-(T)ime Database
-===========================
+============================
 .. currentmodule:: qlib
 
 
 Introduction
 ------------
-Point-in-time data is a very important consideration when performing any sort of historical market analysis. 
+Point-in-time data is a very important consideration when performing any sort of historical market analysis.
 
 For example, let’s say we are backtesting a trading strategy and we are using the past five years of historical data as our input.
 Our model is assumed to trade once a day, at the market close, and we’ll say we are calculating the trading signal for 1 January 2020 in our backtest. At that point, we should only have data for 1 January 2020, 31 December 2019, 30 December 2019 etc.

docs/advanced/alpha.rst

@@ -1,12 +1,12 @@
 .. _alpha:
 
-===========================
-Building Formulaic Alphas 
-===========================
+=========================
+Building Formulaic Alphas
+=========================
 .. currentmodule:: qlib
 
 Introduction
-===================
+============
 
 In quantitative trading practice, designing novel factors that can explain and predict future asset returns are of vital importance to the profitability of a strategy. Such factors are usually called alpha factors, or alphas in short.
 
@@ -15,30 +15,30 @@ A formulaic alpha, as the name suggests, is a kind of alpha that can be presente
 
 
 Building Formulaic Alphas in ``Qlib``
-======================================
+=====================================
 
 In ``Qlib``, users can easily build formulaic alphas.
 
 Example
------------------
+-------
 
 `MACD`, short for moving average convergence/divergence, is a formulaic alpha used in technical analysis of stock prices. It is designed to reveal changes in the strength, direction, momentum, and duration of a trend in a stock's price.
 
 `MACD` can be presented as the following formula:
 
-.. math:: 
+.. math::
 
     MACD = 2\times (DIF-DEA)
 
 .. note::
 
     `DIF` means Differential value, which is 12-period EMA minus 26-period EMA.
-    
+
     .. math::
 
-        DIF = \frac{EMA(CLOSE, 12) - EMA(CLOSE, 26)}{CLOSE} 
+        DIF = \frac{EMA(CLOSE, 12) - EMA(CLOSE, 26)}{CLOSE}
 
-    `DEA`means a 9-period EMA of the DIF.
+    `DEA` means a 9-period EMA of the DIF.
 
     .. math::
 
@@ -65,7 +65,7 @@ Users can use ``Data Handler`` to build formulaic alphas `MACD` in qlib:
     >> print(df)
                             feature     label
                                MACD     LABEL
-    datetime   instrument                    
+    datetime   instrument
     2010-01-04 SH600000   -0.011547 -0.019672
                SH600004    0.002745 -0.014721
                SH600006    0.010133  0.002911
@@ -79,7 +79,7 @@ Users can use ``Data Handler`` to build formulaic alphas `MACD` in qlib:
                SZ300315   -0.030557  0.012455
 
 Reference
-===========
+=========
 
 To learn more about ``Data Loader``, please refer to `Data Loader <../component/data.html#data-loader>`_
 

docs/advanced/serial.rst

@@ -1,26 +1,26 @@
 .. _serial:
 
-=================================
+=============
 Serialization
-=================================
+=============
 .. currentmodule:: qlib
 
 Introduction
-===================
-``Qlib`` supports dumping the state of ``DataHandler``, ``DataSet``, ``Processor`` and ``Model``, etc. into a disk and reloading them. 
+============
+``Qlib`` supports dumping the state of ``DataHandler``, ``DataSet``, ``Processor`` and ``Model``, etc. into a disk and reloading them.
 
 Serializable Class
-========================
+==================
 
-``Qlib`` provides a base class ``qlib.utils.serial.Serializable``, whose state can be dumped into or loaded from disk in `pickle` format. 
+``Qlib`` provides a base class ``qlib.utils.serial.Serializable``, whose state can be dumped into or loaded from disk in `pickle` format.
 When users dump the state of a ``Serializable`` instance, the attributes of the instance whose name **does not** start with `_` will be saved on the disk.
 However, users can use ``config`` method or override ``default_dump_all`` attribute to prevent this feature.
 
 Users can also override ``pickle_backend`` attribute to choose a pickle backend. The supported value is "pickle" (default and common) and "dill" (dump more things such as function, more information in `here <https://pypi.org/project/dill/>`_).
 
 Example
-==========================
-``Qlib``'s serializable class includes  ``DataHandler``, ``DataSet``, ``Processor`` and ``Model``, etc., which are subclass of  ``qlib.utils.serial.Serializable``. 
+=======
+``Qlib``'s serializable class includes  ``DataHandler``, ``DataSet``, ``Processor`` and ``Model``, etc., which are subclass of  ``qlib.utils.serial.Serializable``.
 Specifically, ``qlib.data.dataset.DatasetH`` is one of them. Users can serialize ``DatasetH`` as follows.
 
 .. code-block:: Python
@@ -33,7 +33,7 @@ Specifically, ``qlib.data.dataset.DatasetH`` is one of them. Users can serialize
         dataset = pickle.load(file_dataset)
 
 .. note::
-    Only state of ``DatasetH`` should be saved on the disk, such as some `mean` and `variance` used for data normalization, etc. 
+    Only state of ``DatasetH`` should be saved on the disk, such as some `mean` and `variance` used for data normalization, etc.
 
     After reloading the ``DatasetH``, users need to reinitialize it. It means that users can reset some states of ``DatasetH`` or ``QlibDataHandler`` such as `instruments`, `start_time`, `end_time` and `segments`, etc.,  and generate new data according to the states (data is not state and should not be saved on the disk).
 
@@ -41,5 +41,5 @@ A more detailed example is in this `link <https://github.com/microsoft/qlib/tree
 
 
 API
-===================
+===
 Please refer to `Serializable API <../reference/api.html#module-qlib.utils.serial.Serializable>`_.

docs/advanced/server.rst

@@ -1,15 +1,15 @@
 .. _server:
 
-=================================
+=============================
 ``Online`` & ``Offline`` mode
-=================================
+=============================
 .. currentmodule:: qlib
 
 
 Introduction
-=============
+============
 
-``Qlib`` supports ``Online`` mode and ``Offline`` mode. Only the ``Offline`` mode is introduced in this document. 
+``Qlib`` supports ``Online`` mode and ``Offline`` mode. Only the ``Offline`` mode is introduced in this document.
 
 The ``Online`` mode is designed to solve the following problems:
 
@@ -18,12 +18,12 @@ The ``Online`` mode is designed to solve the following problems:
 - Make the data can be accessed in a remote way.
 
 Qlib-Server
-===============
+===========
 
-``Qlib-Server`` is the assorted server system for ``Qlib``, which utilizes ``Qlib`` for basic calculations and provides extensive server system and cache mechanism. With QLibServer, the data provided for ``Qlib`` can be managed in a centralized manner. With ``Qlib-Server``, users can use ``Qlib`` in ``Online`` mode. 
+``Qlib-Server`` is the assorted server system for ``Qlib``, which utilizes ``Qlib`` for basic calculations and provides extensive server system and cache mechanism. With QLibServer, the data provided for ``Qlib`` can be managed in a centralized manner. With ``Qlib-Server``, users can use ``Qlib`` in ``Online`` mode.
 
 
 
 Reference
-=================
-If users are interested in ``Qlib-Server`` and ``Online`` mode, please refer to `Qlib-Server Project <https://github.com/microsoft/qlib-server>`_ and `Qlib-Server Document <https://qlib-server.readthedocs.io/en/latest/>`_.
+=========
+If users are interested in ``Qlib-Server`` and ``Online`` mode, please refer to `Qlib-Server Project <https://github.com/microsoft/qlib-server>`_ and `Qlib-Server Document <https://qlib-server.readthedocs.io/en/latest/>`_.

docs/advanced/task_management.rst

@@ -1,13 +1,13 @@
 .. _task_management:
 
-=================================
+===============
 Task Management
-=================================
+===============
 .. currentmodule:: qlib
 
 
 Introduction
-=============
+============
 
 The `Workflow <../component/introduction.html>`_ part introduces how to run research workflow in a loosely-coupled way. But it can only execute one ``task`` when you use ``qrun``.
 To automatically generate and execute different tasks, ``Task Management`` provides a whole process including `Task Generating`_, `Task Storing`_, `Task Training`_ and `Task Collecting`_. 
@@ -18,7 +18,7 @@ With this module, users can run their ``task`` automatically at different period
 
 This whole process can be used in `Online Serving <../component/online.html>`_.
 
-An example of the entire process is shown `here <https://github.com/microsoft/qlib/tree/main/examples/model_rolling/task_manager_rolling.py>`_.
+An example of the entire process is shown `here <https://github.com/microsoft/qlib/tree/main/examples/model_rolling/task_manager_rolling.py>`__.
 
 Task Generating
 ===============
@@ -31,12 +31,13 @@ Here is the base class of ``TaskGen``:
 
 .. autoclass:: qlib.workflow.task.gen.TaskGen
     :members:
+    :noindex:
 
 ``Qlib`` provides a class `RollingGen <https://github.com/microsoft/qlib/tree/main/qlib/workflow/task/gen.py>`_ to generate a list of ``task`` of the dataset in different date segments.
-This class allows users to verify the effect of data from different periods on the model in one experiment. More information is `here <../reference/api.html#TaskGen>`_.
+This class allows users to verify the effect of data from different periods on the model in one experiment. More information is `here <../reference/api.html#TaskGen>`__.
 
 Task Storing
-===============
+============
 To achieve higher efficiency and the possibility of cluster operation, ``Task Manager`` will store all tasks in `MongoDB <https://www.mongodb.com/>`_.
 ``TaskManager`` can fetch undone tasks automatically and manage the lifecycle of a set of tasks with error handling.
 Users **MUST** finish the configuration of `MongoDB <https://www.mongodb.com/>`_ when using this module.
@@ -53,22 +54,25 @@ Users need to provide the MongoDB URL and database name for using ``TaskManager`
 
 .. autoclass:: qlib.workflow.task.manage.TaskManager
     :members:
+    :noindex:
 
-More information of ``Task Manager`` can be found in `here <../reference/api.html#TaskManager>`_.
+More information of ``Task Manager`` can be found in `here <../reference/api.html#TaskManager>`__.
 
 Task Training
-===============
+=============
 After generating and storing those ``task``, it's time to run the ``task`` which is in the *WAITING* status.
 ``Qlib`` provides a method called ``run_task`` to run those ``task`` in task pool, however, users can also customize how tasks are executed.
 An easy way to get the ``task_func`` is using ``qlib.model.trainer.task_train`` directly.
 It will run the whole workflow defined by ``task``, which includes *Model*, *Dataset*, *Record*.
 
 .. autofunction:: qlib.workflow.task.manage.run_task
+    :noindex:
 
 Meanwhile, ``Qlib`` provides a module called ``Trainer``. 
 
 .. autoclass:: qlib.model.trainer.Trainer
     :members:
+    :noindex:
 
 ``Trainer`` will train a list of tasks and return a list of model recorders.
 ``Qlib`` offer two kinds of Trainer, TrainerR is the simplest way and TrainerRM is based on TaskManager to help manager tasks lifecycle automatically. 

docs/changelog/changelog.rst

@@ -1,2 +1 @@
 .. include:: ../../CHANGES.rst
-	     

docs/component/data.rst

@@ -1,13 +1,13 @@
 .. _data:
 
-================================
+==================================
 Data Layer: Data Framework & Usage
-================================
+==================================
 
 Introduction
-============================
+============
 
-``Data Layer`` provides user-friendly APIs to manage and retrieve data. It provides high-performance data infrastructure. 
+``Data Layer`` provides user-friendly APIs to manage and retrieve data. It provides high-performance data infrastructure.
 
 It is designed for quantitative investment. For example, users could build formulaic alphas with ``Data Layer`` easily. Please refer to `Building Formulaic Alphas <../advanced/alpha.html>`_ for more details.
 
@@ -23,21 +23,21 @@ The introduction of ``Data Layer`` includes the following parts.
 
 Here is a typical example of Qlib data workflow
 
-- Users download data and converting data into Qlib format(with filename suffix `.bin`).  In this step, typically only some basic data are stored on disk(such as OHLCV). 
-- Creating some basic features based on Qlib's expression Engine(e.g. "Ref($close, 60) / $close", the return of last 60 trading days). Supported operators in the expression engine can be found `here <https://github.com/microsoft/qlib/blob/main/qlib/data/ops.py>`_. This step is typically implemented in Qlib's `Data Loader <https://qlib.readthedocs.io/en/latest/component/data.html#data-loader>`_ which is a component of `Data Handler <https://qlib.readthedocs.io/en/latest/component/data.html#data-handler>`_ .
-- If users require more complicated data processing (e.g. data normalization),  `Data Handler <https://qlib.readthedocs.io/en/latest/component/data.html#data-handler>`_ support user-customized processors to process data(some predefined processors can be found `here <https://github.com/microsoft/qlib/blob/main/qlib/data/dataset/processor.py>`_).  The processors are different from operators in expression engine. It is designed for some complicated data processing methods which is hard to supported in operators in expression engine.
+- Users download data and converting data into Qlib format(with filename suffix `.bin`).  In this step, typically only some basic data are stored on disk(such as OHLCV).
+- Creating some basic features based on Qlib's expression Engine(e.g. "Ref($close, 60) / $close", the return of last 60 trading days). Supported operators in the expression engine can be found `here <https://github.com/microsoft/qlib/blob/main/qlib/data/ops.py>`__. This step is typically implemented in Qlib's `Data Loader <https://qlib.readthedocs.io/en/latest/component/data.html#data-loader>`_ which is a component of `Data Handler <https://qlib.readthedocs.io/en/latest/component/data.html#data-handler>`_ .
+- If users require more complicated data processing (e.g. data normalization),  `Data Handler <https://qlib.readthedocs.io/en/latest/component/data.html#data-handler>`_ support user-customized processors to process data(some predefined processors can be found `here <https://github.com/microsoft/qlib/blob/main/qlib/data/dataset/processor.py>`__).  The processors are different from operators in expression engine. It is designed for some complicated data processing methods which is hard to supported in operators in expression engine.
 - At last, `Dataset <https://qlib.readthedocs.io/en/latest/component/data.html#dataset>`_ is responsible to prepare model-specific dataset from the processed data of Data Handler
 
 Data Preparation
-============================
+================
 
 Qlib Format Data
-------------------
+----------------
 
 We've specially designed a data structure to manage financial data, please refer to the `File storage design section in Qlib paper <https://arxiv.org/abs/2009.11189>`_ for detailed information.
 Such data will be stored with filename suffix `.bin` (We'll call them `.bin` file, `.bin` format, or qlib format). `.bin` file is designed for scientific computing on finance data.
 
-``Qlib`` provides two different off-the-shelf datasets, which can be accessed through this `link <https://github.com/microsoft/qlib/blob/main/qlib/contrib/data/handler.py>`_:
+``Qlib`` provides two different off-the-shelf datasets, which can be accessed through this `link <https://github.com/microsoft/qlib/blob/main/qlib/contrib/data/handler.py>`__:
 
 ========================  =================  ================
 Dataset                   US Market          China Market
@@ -47,14 +47,19 @@ Alpha360                  √                  √
 Alpha158                  √                  √
 ========================  =================  ================
 
-Also, ``Qlib`` provides a high-frequency dataset. Users can run a high-frequency dataset example through this `link <https://github.com/microsoft/qlib/tree/main/examples/highfreq>`_.
+Also, ``Qlib`` provides a high-frequency dataset. Users can run a high-frequency dataset example through this `link <https://github.com/microsoft/qlib/tree/main/examples/highfreq>`__.
 
 Qlib Format Dataset
---------------------
-``Qlib`` has provided an off-the-shelf dataset in `.bin` format, users could use the script ``scripts/get_data.py`` to download the China-Stock dataset as follows.
-The price volume data look different from the actual dealling price because of they are **adjusted** (`adjusted price <https://www.investopedia.com/terms/a/adjusted_closing_price.asp>`_).  And then you may find that the adjusted price may be different from different data sources. This is because different data sources may vary in the way of adjusting prices. Qlib normalize the price on first trading day of each stock to 1 when adjusting them. 
+-------------------
+``Qlib`` has provided an off-the-shelf dataset in `.bin` format, users could use the script ``scripts/get_data.py`` to download the China-Stock dataset as follows. User can also use numpy to load `.bin` file to validate data.
+The price volume data look different from the actual dealling price because of they are **adjusted** (`adjusted price <https://www.investopedia.com/terms/a/adjusted_closing_price.asp>`_).  And then you may find that the adjusted price may be different from different data sources. This is because different data sources may vary in the way of adjusting prices. Qlib normalize the price on first trading day of each stock to 1 when adjusting them.
 Users can leverage `$factor` to get the original trading price (e.g. `$close / $factor` to get the original close price).
 
+Here are some discussions about the price adjusting of Qlib. 
+
+- https://github.com/microsoft/qlib/issues/991#issuecomment-1075252402
+
+
 .. code-block:: bash
 
     # download 1d
@@ -104,7 +109,7 @@ Automatic update of daily frequency data
 
 
 Converting CSV Format into Qlib Format
--------------------------------------------
+--------------------------------------
 
 ``Qlib`` has provided the script ``scripts/dump_bin.py`` to convert **any** data in CSV format into `.bin` files (``Qlib`` format) as long as they are in the correct format.
 
@@ -126,16 +131,16 @@ Users can also provide their own data in CSV format. However, the CSV data **mus
 - CSV file is named after a specific stock *or* the CSV file includes a column of the stock name
 
     - Name the CSV file after a stock: `SH600000.csv`, `AAPL.csv` (not case sensitive).
-    
+
     - CSV file includes a column of the stock name. User **must** specify the column name when dumping the data. Here is an example:
 
         .. code-block:: bash
 
             python scripts/dump_bin.py dump_all ... --symbol_field_name symbol
-        
+
         where the data are in the following format:
 
-        .. code-block:: 
+        .. code-block::
 
             symbol,close
             SH600000,120
@@ -145,10 +150,10 @@ Users can also provide their own data in CSV format. However, the CSV data **mus
     .. code-block:: bash
 
         python scripts/dump_bin.py dump_all ... --date_field_name date
-    
+
     where the data are in the following format:
 
-    .. code-block:: 
+    .. code-block::
 
         symbol,date,close,open,volume
         SH600000,2020-11-01,120,121,12300000
@@ -172,7 +177,7 @@ After conversion, users can find their Qlib format data in the directory `~/.qli
 .. note::
 
     The arguments of `--include_fields` should correspond with the column names of CSV files. The columns names of dataset provided by ``Qlib`` should include open, close, high, low, volume and factor at least.
-    
+
     - `open`
         The adjusted opening price
     - `close`
@@ -186,11 +191,11 @@ After conversion, users can find their Qlib format data in the directory `~/.qli
     - `factor`
         The Restoration factor. Normally, ``factor = adjusted_price / original_price``, `adjusted price` reference: `split adjusted <https://www.investopedia.com/terms/s/splitadjusted.asp>`_
 
-    In the convention of `Qlib` data processing, `open, close, high, low, volume, money and factor` will be set to NaN if the stock is suspended. 
+    In the convention of `Qlib` data processing, `open, close, high, low, volume, money and factor` will be set to NaN if the stock is suspended.
     If you want to use your own alpha-factor which can't be calculate by OCHLV, like PE, EPS and so on, you could add it to the CSV files with OHCLV together and then dump it to the Qlib format data.
 
 Stock Pool (Market)
---------------------------------
+-------------------
 
 ``Qlib`` defines `stock pool <https://github.com/microsoft/qlib/blob/main/examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml#L4>`_ as stock list and their date ranges. Predefined stock pools (e.g. csi300) may be imported as follows.
 
@@ -200,7 +205,7 @@ Stock Pool (Market)
 
 
 Multiple Stock Modes
---------------------------------
+--------------------
 
 ``Qlib`` now provides two different stock modes for users: China-Stock Mode & US-Stock Mode. Here are some different settings of these two modes:
 
@@ -218,23 +223,23 @@ The `trade unit` defines the unit number of stocks can be used in a trade, and t
     - Download china-stock in qlib format, please refer to section `Qlib Format Dataset <#qlib-format-dataset>`_.
     - Initialize ``Qlib`` in china-stock mode
         Supposed that users download their Qlib format data in the directory ``~/.qlib/qlib_data/cn_data``. Users only need to initialize ``Qlib`` as follows.
-        
+
         .. code-block:: python
 
             from qlib.constant import REG_CN
             qlib.init(provider_uri='~/.qlib/qlib_data/cn_data', region=REG_CN)
-        
+
 
 - If users use ``Qlib`` in US-stock mode, US-stock data is required. ``Qlib`` also provides a script to download US-stock data. Users can use ``Qlib`` in US-stock mode according to the following steps:
     - Download us-stock in qlib format, please refer to section `Qlib Format Dataset <#qlib-format-dataset>`_.
     - Initialize ``Qlib`` in US-stock mode
         Supposed that users prepare their Qlib format data in the directory ``~/.qlib/qlib_data/us_data``. Users only need to initialize ``Qlib`` as follows.
-        
+
         .. code-block:: python
 
             from qlib.config import REG_US
             qlib.init(provider_uri='~/.qlib/qlib_data/us_data', region=REG_US)
-        
+
 
 .. note::
 
@@ -242,14 +247,14 @@ The `trade unit` defines the unit number of stocks can be used in a trade, and t
 
 
 Data API
-========================
+========
 
 Data Retrieval
----------------
+--------------
 Users can use APIs in ``qlib.data`` to retrieve data, please refer to `Data Retrieval <../start/getdata.html>`_.
 
 Feature
-------------------
+-------
 
 ``Qlib`` provides `Feature` and `ExpressionOps` to fetch the features according to users' needs.
 
@@ -264,7 +269,7 @@ Feature
 To know more about  ``Feature``, please refer to `Feature API <../reference/api.html#module-qlib.data.base>`_.
 
 Filter
--------------------
+------
 ``Qlib`` provides `NameDFilter` and `ExpressionDFilter` to filter the instruments according to users' needs.
 
 - `NameDFilter`
@@ -272,7 +277,7 @@ Filter
 
 - `ExpressionDFilter`
     Expression dynamic instrument filter. Filter the instruments based on a certain expression. An expression rule indicating a certain feature field is required.
-    
+
     - `basic features filter`: rule_expression = '$close/$open>5'
     - `cross-sectional features filter` \: rule_expression = '$rank($close)<10'
     - `time-sequence features filter`: rule_expression = '$Ref($close, 3)>100'
@@ -299,63 +304,65 @@ Here is a simple example showing how to use filter in a basic ``Qlib`` workflow
 To know more about ``Filter``, please refer to `Filter API <../reference/api.html#module-qlib.data.filter>`_.
 
 Reference
--------------
+---------
 
 To know more about ``Data API``, please refer to `Data API <../reference/api.html#data>`_.
 
 
 Data Loader
-=================
+===========
 
 ``Data Loader`` in ``Qlib`` is designed to load raw data from the original data source. It will be loaded and used in the ``Data Handler`` module.
 
 QlibDataLoader
----------------
+--------------
 
 The ``QlibDataLoader`` class in ``Qlib`` is such an interface that allows users to load raw data from the ``Qlib`` data source.
 
 StaticDataLoader
----------------
+----------------
 
 The ``StaticDataLoader`` class in ``Qlib`` is such an interface that allows users to load raw data from file or as provided.
 
 
 Interface
-------------
+---------
 
 Here are some interfaces of the ``QlibDataLoader`` class:
 
 .. autoclass:: qlib.data.dataset.loader.DataLoader
     :members:
+    :noindex:
 
 API
------------
+---
 
 To know more about ``Data Loader``, please refer to `Data Loader API <../reference/api.html#module-qlib.data.dataset.loader>`_.
 
 
 Data Handler
-=================
+============
 
 The ``Data Handler`` module in ``Qlib`` is designed to handler those common data processing methods which will be used by most of the models.
 
-Users can use ``Data Handler`` in an automatic workflow by ``qrun``, refer to `Workflow: Workflow Management <workflow.html>`_ for more details. 
+Users can use ``Data Handler`` in an automatic workflow by ``qrun``, refer to `Workflow: Workflow Management <workflow.html>`_ for more details.
 
 DataHandlerLP
---------------
+-------------
 
-In addition to use ``Data Handler`` in an automatic workflow with ``qrun``, ``Data Handler`` can be used as an independent module, by which users can easily preprocess data (standardization, remove NaN, etc.) and build datasets. 
+In addition to use ``Data Handler`` in an automatic workflow with ``qrun``, ``Data Handler`` can be used as an independent module, by which users can easily preprocess data (standardization, remove NaN, etc.) and build datasets.
 
 In order to achieve so, ``Qlib`` provides a base class `qlib.data.dataset.DataHandlerLP <../reference/api.html#qlib.data.dataset.handler.DataHandlerLP>`_. The core idea of this class is that: we will have some learnable ``Processors`` which can learn the parameters of data processing(e.g., parameters for zscore normalization). When new data comes in, these `trained` ``Processors`` can then process the new data and thus processing real-time data in an efficient way becomes possible. More information about ``Processors`` will be listed in the next subsection.
 
 
 Interface
-----------------------
+---------
 
 Here are some important interfaces that ``DataHandlerLP`` provides:
 
 .. autoclass:: qlib.data.dataset.handler.DataHandlerLP
     :members: __init__, fetch, get_cols
+    :noindex:
 
 
 If users want to load features and labels by config, users can define a new handler and call the static method `parse_config_to_fields` of ``qlib.contrib.data.handler.Alpha158``.
@@ -364,7 +371,7 @@ Also, users can pass ``qlib.contrib.data.processor.ConfigSectionProcessor`` that
 
 
 Processor
-----------
+---------
 
 The ``Processor`` module in ``Qlib`` is designed to be learnable and it is responsible for handling data processing such as `normalization` and `drop none/nan features/labels`.
 
@@ -382,14 +389,14 @@ The ``Processor`` module in ``Qlib`` is designed to be learnable and it is respo
 - ``CSRankNorm``: `processor` that applies cross sectional rank normalization.
 - ``CSZFillna``: `processor` that fills N/A values in a cross sectional way by the mean of the column.
 
-Users can also create their own `processor` by inheriting the base class of ``Processor``. Please refer to the implementation of all the processors for more information (`Processor Link <https://github.com/microsoft/qlib/blob/main/qlib/data/dataset/processor.py>`_). 
+Users can also create their own `processor` by inheriting the base class of ``Processor``. Please refer to the implementation of all the processors for more information (`Processor Link <https://github.com/microsoft/qlib/blob/main/qlib/data/dataset/processor.py>`_).
 
 To know more about ``Processor``, please refer to `Processor API <../reference/api.html#module-qlib.data.dataset.processor>`_.
 
 Example
---------------
+-------
 
-``Data Handler`` can be run with ``qrun`` by modifying the configuration file, and can also be used as a single module. 
+``Data Handler`` can be run with ``qrun`` by modifying the configuration file, and can also be used as a single module.
 
 Know more about how to run ``Data Handler`` with ``qrun``, please refer to `Workflow: Workflow Management <workflow.html>`_
 
@@ -427,17 +434,17 @@ Qlib provides implemented data handler `Alpha158`. The following example shows h
 .. note:: In the ``Alpha158``, ``Qlib`` uses the label `Ref($close, -2)/Ref($close, -1) - 1` that means the change from T+1 to T+2, rather than `Ref($close, -1)/$close - 1`, of which the reason is that when getting the T day close price of a china stock, the stock can be bought on T+1 day and sold on T+2 day.
 
 API
----------
+---
 
 To know more about ``Data Handler``, please refer to `Data Handler API <../reference/api.html#module-qlib.data.dataset.handler>`_.
 
 
 Dataset
-=================
+=======
 
 The ``Dataset`` module in ``Qlib`` aims to prepare data for model training and inferencing.
 
-The motivation of this module is that we want to maximize the flexibility of different models to handle data that are suitable for themselves. This module gives the model the flexibility to process their data in an unique way. For instance, models such as ``GBDT`` may work well on data that contains `nan` or `None` value, while neural networks such as ``MLP`` will break down on such data. 
+The motivation of this module is that we want to maximize the flexibility of different models to handle data that are suitable for themselves. This module gives the model the flexibility to process their data in an unique way. For instance, models such as ``GBDT`` may work well on data that contains `nan` or `None` value, while neural networks such as ``MLP`` will break down on such data.
 
 If user's model need process its data in a different way, user could implement his own ``Dataset`` class. If the model's
 data processing is not special, ``DatasetH`` can be used directly.
@@ -446,32 +453,35 @@ The ``DatasetH`` class is the `dataset` with `Data Handler`. Here is the most im
 
 .. autoclass:: qlib.data.dataset.__init__.DatasetH
     :members:
+    :noindex:
 
 API
----------
+---
 
 To know more about ``Dataset``, please refer to `Dataset API <../reference/api.html#dataset>`_.
 
 
 Cache
-==========
+=====
 
 ``Cache`` is an optional module that helps accelerate providing data by saving some frequently-used data as cache file. ``Qlib`` provides a `Memcache` class to cache the most-frequently-used data in memory, an inheritable `ExpressionCache` class, and an inheritable `DatasetCache` class.
 
 Global Memory Cache
----------------------
+-------------------
 
 `Memcache` is a global memory cache mechanism that composes of three `MemCacheUnit` instances to cache **Calendar**, **Instruments**, and **Features**. The `MemCache` is defined globally in `cache.py` as `H`. Users can use `H['c'], H['i'], H['f']` to get/set `memcache`.
 
 .. autoclass:: qlib.data.cache.MemCacheUnit
     :members:
+    :noindex:
 
 .. autoclass:: qlib.data.cache.MemCache
     :members:
+    :noindex:
 
 
 ExpressionCache
------------------
+---------------
 
 `ExpressionCache` is a cache mechanism that saves expressions such as **Mean($close, 5)**. Users can inherit this base class to define their own cache mechanism that saves expressions according to the following steps.
 
@@ -482,11 +492,12 @@ The following shows the details about the interfaces:
 
 .. autoclass:: qlib.data.cache.ExpressionCache
     :members:
+    :noindex:
 
 ``Qlib`` has currently provided implemented disk cache `DiskExpressionCache` which inherits from `ExpressionCache` . The expressions data will be stored in the disk.
 
 DatasetCache
------------------
+------------
 
 `DatasetCache` is a cache mechanism that saves datasets. A certain dataset is regulated by a stock pool configuration (or a series of instruments, though not recommended), a list of expressions or static feature fields, the start time, and end time for the collected features and the frequency. Users can inherit this base class to define their own cache mechanism that saves datasets according to the following steps.
 
@@ -497,17 +508,18 @@ The following shows the details about the interfaces:
 
 .. autoclass:: qlib.data.cache.DatasetCache
     :members:
+    :noindex:
 
 ``Qlib`` has currently provided implemented disk cache `DiskDatasetCache` which inherits from `DatasetCache` . The datasets' data will be stored in the disk.
 
 
 
 Data and Cache File Structure
-==================================
+=============================
 
 We've specially designed a file structure to manage data and cache, please refer to the `File storage design section in Qlib paper <https://arxiv.org/abs/2009.11189>`_ for detailed information. The file structure of data and cache is listed as follows.
 
-.. code-block:: json
+.. code-block::
 
     - data/
         [raw data] updated by data providers
@@ -536,4 +548,3 @@ We've specially designed a file structure to manage data and cache, please refer
                 - .meta : an assorted meta file recording the stockpool config, field names and visit times
                 - .index : an assorted index file recording the line index of all calendars
             - ...
-

docs/component/highfreq.rst

@@ -1,38 +1,40 @@
 .. _highfreq:
 
-============================================
+========================================================================
 Design of Nested Decision Execution Framework for High-Frequency Trading
-============================================
+========================================================================
 .. currentmodule:: qlib
 
 Introduction
-===================
+============
 
-Daily trading (e.g. portfolio management) and intraday trading (e.g. orders execution) are two hot topics in Quant investment and usually studied separately.
+Daily trading (e.g. portfolio management) and intraday trading (e.g. orders execution) are two hot topics in Quant investment and are usually studied separately.
 
 To get the join trading performance of daily and intraday trading, they must interact with each other and run backtest jointly.
-In order to support the joint backtest strategies in multiple levels, a corresponding framework is required. None of the publicly available high-frequency trading frameworks considers multi-level joint trading, which make the backtesting aforementioned inaccurate.
+In order to support the joint backtest strategies at multiple levels, a corresponding framework is required. None of the publicly available high-frequency trading frameworks considers multi-level joint trading, which makes the backtesting aforementioned inaccurate.
 
 Besides backtesting, the optimization of strategies from different levels is not standalone and can be affected by each other.
-For example, the best portfolio management strategy may change with the performance of order executions(e.g. a portfolio with higher turnover may becomes a better choice when we improve the order execution strategies).
-To achieve the overall good performance , it is necessary to consider the interaction of strategies in different level. 
+For example, the best portfolio management strategy may change with the performance of order executions(e.g. a portfolio with higher turnover may become a better choice when we improve the order execution strategies).
+To achieve overall good performance, it is necessary to consider the interaction of strategies at a different levels.
 
-Therefore, building a new framework for trading in multiple levels becomes necessary to solve the various problems mentioned above, for which we designed a nested decision execution framework that consider the interaction of strategies.
+Therefore, building a new framework for trading on multiple levels becomes necessary to solve the various problems mentioned above, for which we designed a nested decision execution framework that considers the interaction of strategies.
 
 .. image:: ../_static/img/framework.svg
 
-The design of the framework is shown in the yellow part in the middle of the figure above. Each level consists of ``Trading Agent`` and ``Execution Env``. ``Trading Agent`` has its own data processing module (``Information Extractor``), forecasting module (``Forecast Model``) and decision generator (``Decision Generator``). The trading algorithm generates the decisions by the ``Decision Generator`` based on the forecast signals output by the ``Forecast Module``, and the decisions generated by the trading algorithm are passed to the ``Execution Env``, which returns the execution results. 
+The design of the framework is shown in the yellow part in the middle of the figure above. Each level consists of ``Trading Agent`` and ``Execution Env``. ``Trading Agent`` has its own data processing module (``Information Extractor``), forecasting module (``Forecast Model``) and decision generator (``Decision Generator``). The trading algorithm generates the decisions by the ``Decision Generator`` based on the forecast signals output by the ``Forecast Module``, and the decisions generated by the trading algorithm are passed to the ``Execution Env``, which returns the execution results.
 
-The frequency of trading algorithm, decision content and execution environment can be customized by users (e.g. intraday trading, daily-frequency trading, weekly-frequency trading), and the execution environment can be nested with finer-grained trading algorithm and execution environment inside (i.e. sub-workflow in the figure, e.g. daily-frequency orders can be turned into finer-grained decisions by splitting orders within the day). The flexibility of nested decision execution framework makes it easy for users to explore the effects of combining different levels of trading strategies and break down the optimization barriers between different levels of trading algorithm.
+The frequency of the trading algorithm, decision content and execution environment can be customized by users (e.g. intraday trading, daily-frequency trading, weekly-frequency trading), and the execution environment can be nested with finer-grained trading algorithm and execution environment inside (i.e. sub-workflow in the figure, e.g. daily-frequency orders can be turned into finer-grained decisions by splitting orders within the day). The flexibility of the nested decision execution framework makes it easy for users to explore the effects of combining different levels of trading strategies and break down the optimization barriers between different levels of the trading algorithm.
+
+The optimization for the nested decision execution framework can be implemented with the support of `QlibRL <https://qlib.readthedocs.io/en/latest/component/rl.html>`_. To know more about how to use the QlibRL, go to API Reference: `RL API <../reference/api.html#rl>`_. 
 
 Example
-===========================
+=======
 
-An example of nested decision execution framework for high-frequency can be found `here <https://github.com/microsoft/qlib/blob/main/examples/nested_decision_execution/workflow.py>`_.
+An example of a nested decision execution framework for high-frequency can be found `here <https://github.com/microsoft/qlib/blob/main/examples/nested_decision_execution/workflow.py>`_.
 
 
-Besides, the above examples, here are some other related work about high-frequency trading in Qlib.
+Besides, the above examples, here are some other related works about high-frequency trading in Qlib.
 
 - `Prediction with high-frequency data <https://github.com/microsoft/qlib/tree/main/examples/highfreq#benchmarks-performance-predicting-the-price-trend-in-high-frequency-data>`_
-- `Examples <https://github.com/microsoft/qlib/blob/main/examples/orderbook_data/>`_ to extract features form high-frequency data without fixed frequency.
+- `Examples <https://github.com/microsoft/qlib/blob/main/examples/orderbook_data/>`_ to extract features from high-frequency data without fixed frequency.
 - `A paper <https://github.com/microsoft/qlib/tree/high-freq-execution#high-frequency-execution>`_ for high-frequency trading.

docs/component/meta.rst

@@ -1,17 +1,17 @@
 .. _meta:
 
-=================================
+======================================================
 Meta Controller: Meta-Task & Meta-Dataset & Meta-Model
-=================================
+======================================================
 .. currentmodule:: qlib
 
 
 Introduction
-=============
+============
 ``Meta Controller`` provides guidance to ``Forecast Model``, which aims to learn regular patterns among a series of forecasting tasks and use learned patterns to guide forthcoming forecasting tasks. Users can implement their own meta-model instance based on ``Meta Controller`` module.
 
 Meta Task
-=============
+=========
 
 A `Meta Task` instance is the basic element in the meta-learning framework. It saves the data that can be used for the `Meta Model`. Multiple `Meta Task` instances may share the same `Data Handler`, controlled by `Meta Dataset`. Users should use `prepare_task_data()` to obtain the data that can be directly fed into the `Meta Model`.
 
@@ -19,7 +19,7 @@ A `Meta Task` instance is the basic element in the meta-learning framework. It s
     :members:
 
 Meta Dataset
-=============
+============
 
 `Meta Dataset` controls the meta-information generating process. It is on the duty of providing data for training the `Meta Model`. Users should use `prepare_tasks` to retrieve a list of `Meta Task` instances.
 
@@ -27,26 +27,26 @@ Meta Dataset
     :members:
 
 Meta Model
-=============
+==========
 
 General Meta Model
 ------------------
 `Meta Model` instance is the part that controls the workflow. The usage of the `Meta Model` includes:
-1. Users train their `Meta Model` with the `fit` function. 
+1. Users train their `Meta Model` with the `fit` function.
 2. The `Meta Model` instance guides the workflow by giving useful information via the `inference` function.
 
 .. autoclass:: qlib.model.meta.model.MetaModel
     :members:
 
 Meta Task Model
-------------------
+---------------
 This type of meta-model may interact with task definitions directly. Then, the `Meta Task Model` is the class for them to inherit from. They guide the base tasks by modifying the base task definitions. The function `prepare_tasks` can be used to obtain the modified base task definitions.
 
 .. autoclass:: qlib.model.meta.model.MetaTaskModel
     :members:
 
 Meta Guide Model
-------------------
+----------------
 This type of meta-model participates in the training process of the base forecasting model. The meta-model may guide the base forecasting models during their training to improve their performances.
 
 .. autoclass:: qlib.model.meta.model.MetaGuideModel
@@ -54,9 +54,9 @@ This type of meta-model participates in the training process of the base forecas
 
 
 Example
-=============
-``Qlib`` provides an implementation of ``Meta Model`` module, ``DDG-DA``, 
-which adapts to the market dynamics. 
+=======
+``Qlib`` provides an implementation of ``Meta Model`` module, ``DDG-DA``,
+which adapts to the market dynamics.
 
 ``DDG-DA`` includes four steps:
 

docs/component/model.rst

@@ -1,13 +1,13 @@
 .. _model:
 
-============================================
+===========================================
 Forecast Model: Model Training & Prediction
-============================================
+===========================================
 
 Introduction
-===================
+============
 
-``Forecast Model`` is designed to make the `prediction score` about stocks. Users can use the ``Forecast Model`` in an automatic workflow by ``qrun``, please refer to `Workflow: Workflow Management <workflow.html>`_.  
+``Forecast Model`` is designed to make the `prediction score` about stocks. Users can use the ``Forecast Model`` in an automatic workflow by ``qrun``, please refer to `Workflow: Workflow Management <workflow.html>`_.
 
 Because the components in ``Qlib`` are designed in a loosely-coupled way, ``Forecast Model`` can be used as an independent module also.
 
@@ -20,13 +20,14 @@ The base class provides the following interfaces:
 
 .. autoclass:: qlib.model.base.Model
     :members:
+    :noindex:
 
 ``Qlib`` also provides a base class `qlib.model.base.ModelFT <../reference/api.html#qlib.model.base.ModelFT>`_, which includes the method for finetuning the model.
-    
+
 For other interfaces such as `finetune`, please refer to `Model API <../reference/api.html#module-qlib.model.base>`_.
 
 Example
-==================
+=======
 
 ``Qlib``'s `Model Zoo` includes models such as ``LightGBM``, ``MLP``, ``LSTM``, etc.. These models are treated as the baselines of ``Forecast Model``. The following steps show how to run`` LightGBM`` as an independent module.
 
@@ -84,7 +85,7 @@ Example
                 },
             },
         }
-        
+
         # model initiaiton
         model = init_instance_by_config(task["model"])
         dataset = init_instance_by_config(task["dataset"])
@@ -100,22 +101,22 @@ Example
             sr = SignalRecord(model, dataset, recorder)
             sr.generate()
 
-    .. note:: 
-        
+    .. note::
+
         `Alpha158` is the data handler provided by ``Qlib``, please refer to `Data Handler <data.html#data-handler>`_.
         `SignalRecord` is the `Record Template` in ``Qlib``, please refer to `Workflow <recorder.html#record-template>`_.
 
 Also, the above example has been given in ``examples/train_backtest_analyze.ipynb``.
 Technically, the meaning of the model prediction depends on the label setting designed by user.
-By default, the meaning of the score is normally the rating of the instruments by the forecasting model. The higher the score, the more profit the instruments. 
+By default, the meaning of the score is normally the rating of the instruments by the forecasting model. The higher the score, the more profit the instruments.
 
 
 Custom Model
-===================
+============
 
 Qlib supports custom models. If users are interested in customizing their own models and integrating the models into ``Qlib``, please refer to `Custom Model Integration <../start/integration.html>`_.
 
 
 API
-===================
+===
 Please refer to `Model API <../reference/api.html#module-qlib.model.base>`_.

docs/component/online.rst

@@ -1,13 +1,13 @@
-.. _online:
+.. _online_serving:
 
-=================================
+==============
 Online Serving
-=================================
+==============
 .. currentmodule:: qlib
 
 
 Introduction
-=============
+============
 
 .. image:: ../_static/img/online_serving.png
     :align: center
@@ -15,7 +15,7 @@ Introduction
 
 In addition to backtesting, one way to test a model is effective is to make predictions in real market conditions or even do real trading based on those predictions.
 ``Online Serving`` is a set of modules for online models using the latest data,
-which including `Online Manager <#Online Manager>`_, `Online Strategy <#Online Strategy>`_, `Online Tool <#Online Tool>`_, `Updater <#Updater>`_. 
+which including `Online Manager <#Online Manager>`_, `Online Strategy <#Online Strategy>`_, `Online Tool <#Online Tool>`_, `Updater <#Updater>`_.
 
 `Here <https://github.com/microsoft/qlib/tree/main/examples/online_srv>`_ are several examples for reference, which demonstrate different features of ``Online Serving``.
 If you have many models or `task` needs to be managed, please consider `Task Management <../advanced/task_management.html>`_.
@@ -28,25 +28,29 @@ Known limitations currently
 
 
 Online Manager
-=============
+==============
 
 .. automodule:: qlib.workflow.online.manager
     :members:
+    :noindex:
 
 Online Strategy
-=============
+===============
 
 .. automodule:: qlib.workflow.online.strategy
     :members:
+    :noindex:
 
 Online Tool
-=============
+===========
 
 .. automodule:: qlib.workflow.online.utils
     :members:
+    :noindex:
 
 Updater
-=============
+=======
 
 .. automodule:: qlib.workflow.online.update
     :members:
+    :noindex:

docs/component/recorder.rst

@@ -6,8 +6,8 @@ Qlib Recorder: Experiment Management
 .. currentmodule:: qlib
 
 Introduction
-===================
-``Qlib`` contains an experiment management system named ``QlibRecorder``, which is designed to help users handle experiment and analyse results in an efficient way. 
+============
+``Qlib`` contains an experiment management system named ``QlibRecorder``, which is designed to help users handle experiment and analyse results in an efficient way.
 
 There are three components of the system:
 
@@ -34,13 +34,13 @@ Here is a general view of the structure of the system:
             - Recorder 2
             - ...
         - ...
-        
-This experiment management system defines a set of interface and provided a concrete implementation ``MLflowExpManager``, which is based on the machine learning platform: ``MLFlow`` (`link <https://mlflow.org/>`_). 
+
+This experiment management system defines a set of interface and provided a concrete implementation ``MLflowExpManager``, which is based on the machine learning platform: ``MLFlow`` (`link <https://mlflow.org/>`_).
 
 If users set the implementation of ``ExpManager`` to be ``MLflowExpManager``, they can use the command `mlflow ui` to visualize and check the experiment results. For more information, please refer to the related documents `here <https://www.mlflow.org/docs/latest/cli.html#mlflow-ui>`_.
 
 Qlib Recorder
-===================
+=============
 ``QlibRecorder`` provides a high level API for users to use the experiment management system. The interfaces are wrapped in the variable ``R`` in ``Qlib``, and users can directly use ``R`` to interact with the system. The following command shows how to import ``R`` in Python:
 
 .. code-block:: Python
@@ -55,29 +55,31 @@ Here are the available interfaces of ``QlibRecorder``:
     :members:
 
 Experiment Manager
-===================
+==================
 
 The ``ExpManager`` module in ``Qlib`` is responsible for managing different experiments. Most of the APIs of ``ExpManager`` are similar to ``QlibRecorder``, and the most important API will be the ``get_exp`` method. User can directly refer to the documents above for some detailed information about how to use the ``get_exp`` method.
 
 .. autoclass:: qlib.workflow.expm.ExpManager
     :members: get_exp, list_experiments
+    :noindex:
 
 For other interfaces such as `create_exp`, `delete_exp`, please refer to `Experiment Manager API <../reference/api.html#experiment-manager>`_.
 
 Experiment
-===================
+==========
 
 The ``Experiment`` class is solely responsible for a single experiment, and it will handle any operations that are related to an experiment. Basic methods such as `start`, `end` an experiment are included. Besides, methods related to `recorders` are also available: such methods include `get_recorder` and `list_recorders`.
 
 .. autoclass:: qlib.workflow.exp.Experiment
     :members: get_recorder, list_recorders
+    :noindex:
 
 For other interfaces such as `search_records`, `delete_recorder`, please refer to `Experiment API <../reference/api.html#experiment>`_.
 
 ``Qlib`` also provides a default ``Experiment``, which will be created and used under certain situations when users use the APIs such as `log_metrics` or `get_exp`. If the default ``Experiment`` is used, there will be related logged information when running ``Qlib``. Users are able to change the name of the default ``Experiment`` in the config file of ``Qlib`` or during ``Qlib``'s `initialization <../start/initialization.html#parameters>`_, which is set to be '`Experiment`'.
 
 Recorder
-===================
+========
 
 The ``Recorder`` class is responsible for a single recorder. It will handle some detailed operations such as ``log_metrics``, ``log_params`` of a single run. It is designed to help user to easily track results and things being generated during a run.
 
@@ -85,11 +87,12 @@ Here are some important APIs that are not included in the ``QlibRecorder``:
 
 .. autoclass:: qlib.workflow.recorder.Recorder
     :members: list_artifacts, list_metrics, list_params, list_tags
+    :noindex:
 
 For other interfaces such as `save_objects`, `load_object`, please refer to `Recorder API <../reference/api.html#recorder>`_.
 
 Record Template
-===================
+===============
 
 The ``RecordTemp`` class is a class that enables generate experiment results such as IC and backtest in a certain format. We have provided three different `Record Template` class:
 
@@ -107,7 +110,7 @@ Here is a simple example of what is done in ``SigAnaRecord``, which users can re
 
 - ``PortAnaRecord``: This class generates the results of `backtest`. The detailed information about `backtest` as well as the available `strategy`, users can refer to `Strategy <../component/strategy.html>`_ and `Backtest <../component/backtest.html>`_.
 
-Here is a simple exampke of what is done in ``PortAnaRecord``, which users can refer to if they want to do backtest based on their own prediction and label.
+Here is a simple example of what is done in ``PortAnaRecord``, which users can refer to if they want to do backtest based on their own prediction and label.
 
 .. code-block:: Python
 
@@ -131,7 +134,7 @@ Here is a simple exampke of what is done in ``PortAnaRecord``, which users can r
         "close_cost": 0.0015,
         "min_cost": 5,
     }
-    
+
     strategy = TopkDropoutStrategy(**STRATEGY_CONFIG)
     report_normal, positions_normal = normal_backtest(pred_score, strategy=strategy, **BACKTEST_CONFIG)
 

docs/component/report.rst

@@ -1,11 +1,11 @@
 .. _report:
 
-==========================================
+=======================================
 Analysis: Evaluation & Results Analysis
-==========================================
+=======================================
 
 Introduction
-===================
+============
 
 ``Analysis`` is designed to show the graphical reports of ``Intraday Trading`` , which helps users to evaluate and analyse investment portfolios visually. The following are some graphics to view:
 
@@ -24,7 +24,7 @@ All of the accumulated profit metrics(e.g. return, max drawdown) in Qlib are cal
 This avoids the metrics or the plots being skewed exponentially over time.
 
 Graphical Reports
-===================
+=================
 
 Users can run the following code to get all supported reports.
 
@@ -41,16 +41,17 @@ Users can run the following code to get all supported reports.
 
 
 Usage & Example
-===================
+===============
 
 Usage of `analysis_position.report`
 -----------------------------------
 
 API
-~~~~~~~~~~~~~~~~
+~~~
 
 .. automodule:: qlib.contrib.report.analysis_position.report
     :members:
+    :noindex:
 
 Graphical Result
 ~~~~~~~~~~~~~~~~
@@ -58,7 +59,7 @@ Graphical Result
 .. note::
 
     - Axis X: Trading day
-    - Axis Y: 
+    - Axis Y:
         - `cum bench`
             Cumulative returns series of benchmark
         - `cum return wo cost`
@@ -82,34 +83,35 @@ Graphical Result
     - The shaded part above: Maximum drawdown corresponding to `cum return wo cost`
     - The shaded part below: Maximum drawdown corresponding to `cum ex return wo cost`
 
-.. image:: ../_static/img/analysis/report.png 
+.. image:: ../_static/img/analysis/report.png
 
 
 Usage of `analysis_position.score_ic`
 -------------------------------------
 
 API
-~~~~~~~~~~~~~~~~
+~~~
 
 .. automodule:: qlib.contrib.report.analysis_position.score_ic
     :members:
+    :noindex:
 
 
 Graphical Result
-~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~
 
-.. note:: 
+.. note::
 
     - Axis X: Trading day
-    - Axis Y: 
+    - Axis Y:
         - `ic`
             The `Pearson correlation coefficient` series between `label` and `prediction score`.
             In the above example, the `label` is formulated as `Ref($close, -2)/Ref($close, -1)-1`. Please refer to `Data Feature <data.html#feature>`_ for more details.
-                
+
         - `rank_ic`
             The `Spearman's rank correlation coefficient` series between `label` and `prediction score`.
 
-.. image:: ../_static/img/analysis/score_ic.png 
+.. image:: ../_static/img/analysis/score_ic.png
 
 
 .. Usage of `analysis_position.cumulative_return`
@@ -124,7 +126,7 @@ Graphical Result
 .. Graphical Result
 .. ~~~~~~~~~~~~~~~~~
 ..
-.. .. note:: 
+.. .. note::
 ..
 ..     - Axis X: Trading day
 ..     - Axis Y:
@@ -134,27 +136,28 @@ Graphical Result
 ..     - In the **buy_minus_sell** graph, the **y** value of the **weight** graph at the bottom is `buy_weight + sell_weight`.
 ..     - In each graph, the **red line** in the histogram on the right represents the average.
 ..
-.. .. image:: ../_static/img/analysis/cumulative_return_buy.png 
+.. .. image:: ../_static/img/analysis/cumulative_return_buy.png
 ..
-.. .. image:: ../_static/img/analysis/cumulative_return_sell.png 
+.. .. image:: ../_static/img/analysis/cumulative_return_sell.png
 ..
-.. .. image:: ../_static/img/analysis/cumulative_return_buy_minus_sell.png 
+.. .. image:: ../_static/img/analysis/cumulative_return_buy_minus_sell.png
 ..
-.. .. image:: ../_static/img/analysis/cumulative_return_hold.png 
+.. .. image:: ../_static/img/analysis/cumulative_return_hold.png
 
 
 Usage of `analysis_position.risk_analysis`
-----------------------------------------------
+------------------------------------------
 
 API
-~~~~~~~~~~~~~~~~
+~~~
 
 .. automodule:: qlib.contrib.report.analysis_position.risk_analysis
     :members:
+    :noindex:
 
 
 Graphical Result
-~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~
 
 .. note::
 
@@ -174,6 +177,7 @@ Graphical Result
                 The `Information Ratio` without cost.
             - `excess_return_with_cost`
                 The `Information Ratio` with cost.
+
             To know more about `Information Ratio`, please refer to `Information Ratio – IR <https://www.investopedia.com/terms/i/informationratio.asp>`_.
         -  `max_drawdown`
             - `excess_return_without_cost`
@@ -210,7 +214,7 @@ Graphical Result
                     The `Standard Deviation` series of monthly `CAR` (cumulative abnormal return) without cost.
                 - `excess_return_with_cost_max_drawdown`
                     The `Standard Deviation` series of monthly `CAR` (cumulative abnormal return) with cost.
-                
+
 
 .. image:: ../_static/img/analysis/risk_analysis_annualized_return.png
     :align: center
@@ -221,58 +225,59 @@ Graphical Result
 .. image:: ../_static/img/analysis/risk_analysis_information_ratio.png
     :align: center
 
-.. image:: ../_static/img/analysis/risk_analysis_std.png 
+.. image:: ../_static/img/analysis/risk_analysis_std.png
     :align: center
 
 ..
 .. Usage of `analysis_position.rank_label`
-.. ----------------------------------------------
+.. ---------------------------------------
 ..
 .. API
-.. ~~~~~
+.. ~~~
 ..
 .. .. automodule:: qlib.contrib.report.analysis_position.rank_label
 ..     :members:
 ..
 ..
 .. Graphical Result
-.. ~~~~~~~~~~~~~~~~~
+.. ~~~~~~~~~~~~~~~~
 ..
-.. .. note:: 
+.. .. note::
 ..
 ..     - hold/sell/buy graphics:
 ..         - Axis X: Trading day
-..         - Axis Y: 
+..         - Axis Y:
 ..             Average `ranking ratio`of `label` for stocks that is held/sold/bought on the trading day.
 ..
 ..             In the above example, the `label` is formulated as `Ref($close, -1)/$close - 1`. The `ranking ratio` can be formulated as follows.
 ..             .. math::
-..                 
+..
 ..                 ranking\ ratio = \frac{Ascending\ Ranking\ of\ label}{Number\ of\ Stocks\ in\ the\ Portfolio}
 ..
-.. .. image:: ../_static/img/analysis/rank_label_hold.png 
+.. .. image:: ../_static/img/analysis/rank_label_hold.png
 ..     :align: center
 ..
-.. .. image:: ../_static/img/analysis/rank_label_buy.png 
+.. .. image:: ../_static/img/analysis/rank_label_buy.png
 ..     :align: center
 ..
-.. .. image:: ../_static/img/analysis/rank_label_sell.png 
+.. .. image:: ../_static/img/analysis/rank_label_sell.png
 ..     :align: center
 ..
 ..
 
 Usage of `analysis_model.analysis_model_performance`
------------------------------------------------------
+----------------------------------------------------
 
 API
-~~~~~
+~~~
 
 .. automodule:: qlib.contrib.report.analysis_model.analysis_model_performance
     :members:
+    :noindex:
 
 
 Graphical Results
-~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~
 
 .. note::
 
@@ -291,13 +296,13 @@ Graphical Results
             The Difference series between `Cumulative Return` of `Group1` and of `Group5`
         - `long-average`
             The Difference series between `Cumulative Return` of `Group1` and average `Cumulative Return` for all stocks.
-        
+
         The `ranking ratio` can be formulated as follows.
             .. math::
-                
+
                 ranking\ ratio = \frac{Ascending\ Ranking\ of\ label}{Number\ of\ Stocks\ in\ the\ Portfolio}
 
-.. image:: ../_static/img/analysis/analysis_model_cumulative_return.png 
+.. image:: ../_static/img/analysis/analysis_model_cumulative_return.png
     :align: center
 
 .. note::
@@ -305,7 +310,7 @@ Graphical Results
         The distribution of long-short/long-average returns on each trading day
 
 
-.. image:: ../_static/img/analysis/analysis_model_long_short.png 
+.. image:: ../_static/img/analysis/analysis_model_long_short.png
     :align: center
 
 .. TODO: ask xiao yang for detial
@@ -315,14 +320,14 @@ Graphical Results
         - The `Pearson correlation coefficient` series between `labels` and `prediction scores` of stocks in portfolio.
         - The graphics reports can be used to evaluate the `prediction scores`.
 
-.. image:: ../_static/img/analysis/analysis_model_IC.png 
+.. image:: ../_static/img/analysis/analysis_model_IC.png
     :align: center
 
 .. note::
     - Monthly IC
         Monthly average of the `Information Coefficient`
 
-.. image:: ../_static/img/analysis/analysis_model_monthly_IC.png 
+.. image:: ../_static/img/analysis/analysis_model_monthly_IC.png
     :align: center
 
 .. note::
@@ -331,14 +336,14 @@ Graphical Results
     - IC Normal Dist. Q-Q
         The `Quantile-Quantile Plot` is used for the normal distribution of `Information Coefficient` on each trading day.
 
-.. image:: ../_static/img/analysis/analysis_model_NDQ.png 
+.. image:: ../_static/img/analysis/analysis_model_NDQ.png
     :align: center
 
 .. note::
     - Auto Correlation
-        - The `Pearson correlation coefficient` series between the latest `prediction scores` and the `prediction scores` `lag` days ago of stocks in portfolio on each trading day. 
+        - The `Pearson correlation coefficient` series between the latest `prediction scores` and the `prediction scores` `lag` days ago of stocks in portfolio on each trading day.
         - The graphics reports can be used to estimate the turnover rate.
-         
 
-.. image:: ../_static/img/analysis/analysis_model_auto_correlation.png 
+
+.. image:: ../_static/img/analysis/analysis_model_auto_correlation.png
     :align: center

docs/component/rl/framework.rst

@@ -0,0 +1,49 @@
+The Framework of QlibRL
+=======================
+
+QlibRL contains a full set of components that cover the entire lifecycle of an RL pipeline, including building the simulator of the market, shaping states & actions, training policies (strategies), and backtesting strategies in the simulated environment.
+
+QlibRL is basically implemented with the support of Tianshou and Gym frameworks. The high-level structure of QlibRL is demonstrated below:
+
+.. image:: ../../_static/img/QlibRL_framework.png
+   :width: 600
+   :align: center
+
+Here, we briefly introduce each component in the figure.
+
+EnvWrapper
+------------
+EnvWrapper is the complete capsulation of the simulated environment. It receives actions from outside (policy/strategy/agent), simulates the changes in the market, and then replies rewards and updated states, thus forming an interaction loop.
+
+In QlibRL, EnvWrapper is a subclass of gym.Env, so it implements all necessary interfaces of gym.Env. Any classes or pipelines that accept gym.Env should also accept EnvWrapper. Developers do not need to implement their own EnvWrapper to build their own environment. Instead, they only need to implement 4 components of the EnvWrapper:
+
+- `Simulator`
+    The simulator is the core component responsible for the environment simulation. Developers could implement all the logic that is directly related to the environment simulation in the Simulator in any way they like. In QlibRL, there are already two implementations of Simulator for single asset trading: 1) ``SingleAssetOrderExecution``, which is built based on Qlib's backtest toolkits and hence considers a lot of practical trading details but is slow. 2) ``SimpleSingleAssetOrderExecution``, which is built based on a simplified trading simulator, which ignores a lot of details (e.g. trading limitations, rounding) but is quite fast.
+- `State interpreter` 
+    The state interpreter is responsible for "interpret" states in the original format (format provided by the simulator) into states in a format that the policy could understand. For example, transform unstructured raw features into numerical tensors.
+- `Action interpreter` 
+    The action interpreter is similar to the state interpreter. But instead of states, it interprets actions generated by the policy, from the format provided by the policy to the format that is acceptable to the simulator.
+- `Reward function` 
+    The reward function returns a numerical reward to the policy after each time the policy takes an action. 
+
+EnvWrapper will organically organize these components. Such decomposition allows for better flexibility in development. For example, if the developers want to train multiple types of policies in the same environment, they only need to design one simulator and design different state interpreters/action interpreters/reward functions for different types of policies.
+
+QlibRL has well-defined base classes for all these 4 components. All the developers need to do is define their own components by inheriting the base classes and then implementing all interfaces required by the base classes. The API for the above base components can be found `here <../../reference/api.html#module-qlib.rl>`__.
+
+Policy
+------------
+QlibRL directly uses Tianshou's policy. Developers could use policies provided by Tianshou off the shelf, or implement their own policies by inheriting Tianshou's policies.
+
+Training Vessel & Trainer
+-------------------------
+As stated by their names, training vessels and trainers are helper classes used in training. A training vessel is a ship that contains a simulator/interpreters/reward function/policy, and it controls algorithm-related parts of training. Correspondingly, the trainer is responsible for controlling the runtime parts of training.
+
+As you may have noticed, a training vessel itself holds all the required components to build an EnvWrapper rather than holding an instance of EnvWrapper directly. This allows the training vessel to create duplicates of EnvWrapper dynamically when necessary (for example, under parallel training).
+
+With a training vessel, the trainer could finally launch the training pipeline by simple, Scikit-learn-like interfaces (i.e., ``trainer.fit()``).
+
+The API for Trainer and TrainingVessel and can be found `here <../../reference/api.html#module-qlib.rl.trainer>`__.
+
+The RL module is designed in a loosely-coupled way. Currently, RL examples are integrated with concrete business logic.
+But the core part of RL is much simpler than what you see.
+To demonstrate the simple core of RL, `a dedicated notebook <https://github.com/microsoft/qlib/tree/main/examples/rl/simple_example.ipynb>`__ for RL without business loss is created.

docs/component/rl/overall.rst

@@ -0,0 +1,50 @@
+=====================================================
+Reinforcement Learning in Quantitative Trading
+=====================================================
+
+Reinforcement Learning
+======================
+Different from supervised learning tasks such as classification tasks and regression tasks. Another important paradigm in machine learning is Reinforcement Learning, 
+which attempts to optimize an accumulative numerical reward signal by directly interacting with the environment under a few assumptions such as Markov Decision Process(MDP).
+
+As demonstrated in the following figure, an RL system consists of four elements, 1)the agent 2) the environment the agent interacts with 3) the policy that the agent follows to take actions on the environment and 4)the reward signal from the environment to the agent. 
+In general, the agent can perceive and interpret its environment, take actions and learn through reward, to seek long-term and maximum overall reward to achieve an optimal solution.
+
+.. image:: ../../_static/img/RL_framework.png
+   :width: 300
+   :align: center 
+
+RL attempts to learn to produce actions by trial and error. 
+By sampling actions and then observing which one leads to our desired outcome, a policy is obtained to generate optimal actions. 
+In contrast to supervised learning, RL learns this not from a label but from a time-delayed label called a reward. 
+This scalar value lets us know whether the current outcome is good or bad. 
+In a word, the target of RL is to take actions to maximize reward.
+
+The Qlib Reinforcement Learning toolkit (QlibRL) is an RL platform for quantitative investment, which provides support to implement the RL algorithms in Qlib.
+
+
+Potential Application Scenarios in Quantitative Trading
+=======================================================
+RL methods have already achieved outstanding achievement in many applications, such as game playing, resource allocating, recommendation, marketing and advertising, etc.
+Investment is always a continuous process, taking the stock market as an example, investors need to control their positions and stock holdings by one or more buying and selling behaviors, to maximize the investment returns.
+Besides, each buy and sell decision is made by investors after fully considering the overall market information and stock information. 
+From the view of an investor, the process could be described as a continuous decision-making process generated according to interaction with the market, such problems could be solved by the RL algorithms. 
+Following are some scenarios where RL can potentially be used in quantitative investment.
+
+Portfolio Construction
+----------------------
+Portfolio construction is a process of selecting securities optimally by taking a minimum risk to achieve maximum returns. With an RL-based solution, an agent allocates stocks at every time step by obtaining information for each stock and the market. The key is to develop of policy for building a portfolio and make the policy able to pick the optimal portfolio. 
+
+Order Execution
+---------------
+As a fundamental problem in algorithmic trading, order execution aims at fulfilling a specific trading order, either liquidation or acquirement, for a given instrument. Essentially, the goal of order execution is twofold: it not only requires to fulfill the whole order but also targets a more economical execution with maximizing profit gain (or minimizing capital loss). The order execution with only one order of liquidation or acquirement is called single-asset order execution.
+
+Considering stock investment always aim to pursue long-term maximized profits, it usually manifests as a sequential process of continuously adjusting the asset portfolios, execution for multiple orders, including order of liquidation and acquirement, brings more constraints and makes the sequence of execution for different orders should be considered, e.g. before executing an order to buy some stocks, we have to sell at least one stock. The order execution with multiple assets is called multi-asset order execution. 
+
+According to the order execution’s trait of sequential decision-making, an RL-based solution could be applied to solve the order execution. With an RL-based solution, an agent optimizes execution strategy by interacting with the market environment. 
+
+With QlibRL, the RL algorithm in the above scenarios can be easily implemented.
+
+Nested Portfolio Construction and Order Executor
+------------------------------------------------
+QlibRL makes it possible to jointly optimize different levels of strategies/models/agents. Take `Nested Decision Execution Framework <https://github.com/microsoft/qlib/blob/main/examples/nested_decision_execution>`_ as an example, the optimization of order execution strategy and portfolio management strategies can interact with each other to maximize returns.

docs/component/rl/quickstart.rst

@@ -0,0 +1,175 @@
+
+Quick Start
+============
+.. currentmodule:: qlib
+
+QlibRL provides an example of an implementation of a single asset order execution task and the following is an example of the config file to train with QlibRL.
+
+.. code-block:: yaml
+
+    simulator:
+        # Each step contains 30mins
+        time_per_step: 30
+        # Upper bound of volume, should be null or a float between 0 and 1, if it is a float, represent upper bound is calculated by the percentage of the market volume
+        vol_limit: null
+    env:
+        # Concurrent environment workers.
+        concurrency: 1
+        # dummy or subproc or shmem. Corresponding to `parallelism in tianshou <https://tianshou.readthedocs.io/en/master/api/tianshou.env.html#vectorenv>`_.
+        parallel_mode: dummy
+    action_interpreter:
+        class: CategoricalActionInterpreter
+        kwargs:
+            # Candidate actions, it can be a list with length L: [a_1, a_2,..., a_L] or an integer n, in which case the list of length n+1 is auto-generated, i.e., [0, 1/n, 2/n,..., n/n].
+            values: 14
+            # Total number of steps (an upper-bound estimation)
+            max_step: 8
+        module_path: qlib.rl.order_execution.interpreter
+    state_interpreter:
+        class: FullHistoryStateInterpreter
+        kwargs:
+            # Number of dimensions in data.
+            data_dim: 6
+            # Equal to the total number of records. For example, in SAOE per minute, data_ticks is the length of the day in minutes.
+            data_ticks: 240
+            # The total number of steps (an upper-bound estimation). For example, 390min / 30min-per-step = 13 steps.
+            max_step: 8
+            # Provider of the processed data.
+            processed_data_provider:
+                class: PickleProcessedDataProvider
+                module_path: qlib.rl.data.pickle_styled
+                kwargs:
+                    data_dir: ./data/pickle_dataframe/feature
+        module_path: qlib.rl.order_execution.interpreter
+    reward:
+        class: PAPenaltyReward
+        kwargs:
+            # The penalty for a large volume in a short time.
+            penalty: 100.0
+        module_path: qlib.rl.order_execution.reward
+    data:
+        source:
+            order_dir: ./data/training_order_split
+            data_dir: ./data/pickle_dataframe/backtest
+            # number of time indexes
+            total_time: 240
+            # start time index
+            default_start_time: 0
+            # end time index
+            default_end_time: 240
+            proc_data_dim: 6
+        num_workers: 0
+        queue_size: 20
+    network:
+        class: Recurrent
+        module_path: qlib.rl.order_execution.network
+    policy:
+        class: PPO
+        kwargs:
+            lr: 0.0001
+        module_path: qlib.rl.order_execution.policy
+    runtime:
+        seed: 42
+        use_cuda: false
+    trainer:
+        max_epoch: 2
+        # Number of episodes collected in each training iteration
+        repeat_per_collect: 5
+        earlystop_patience: 2
+        # Episodes per collect at training.
+        episode_per_collect: 20
+        batch_size: 16
+        # Perform validation every n iterations
+        val_every_n_epoch: 1
+        checkpoint_path: ./checkpoints
+        checkpoint_every_n_iters: 1
+
+
+And the config file for backtesting:
+
+.. code-block:: yaml
+
+    order_file: ./data/backtest_orders.csv
+    start_time: "9:45"
+    end_time: "14:44"
+    qlib:
+        provider_uri_1min: ./data/bin
+        feature_root_dir: ./data/pickle
+        # feature generated by today's information
+        feature_columns_today: [
+            "$open", "$high", "$low", "$close", "$vwap", "$volume",
+        ]
+        # feature generated by yesterday's information
+        feature_columns_yesterday: [
+            "$open_v1", "$high_v1", "$low_v1", "$close_v1", "$vwap_v1", "$volume_v1",
+        ]
+    exchange:
+        # the expression for buying and selling stock limitation
+        limit_threshold: ['$close == 0', '$close == 0']
+        # deal price for buying and selling
+        deal_price: ["If($close == 0, $vwap, $close)", "If($close == 0, $vwap, $close)"]
+    volume_threshold:
+        # volume limits are both buying and selling, "cum" means that this is a cumulative value over time
+        all: ["cum", "0.2 * DayCumsum($volume, '9:45', '14:44')"]
+        # the volume limits of buying
+        buy: ["current", "$close"]
+        # the volume limits of selling, "current" means that this is a real-time value and will not accumulate over time
+        sell: ["current", "$close"]
+    strategies: 
+        30min: 
+            class: TWAPStrategy
+            module_path: qlib.contrib.strategy.rule_strategy
+            kwargs: {}
+        1day: 
+            class: SAOEIntStrategy
+            module_path: qlib.rl.order_execution.strategy
+            kwargs:
+            state_interpreter:
+                class: FullHistoryStateInterpreter
+                module_path: qlib.rl.order_execution.interpreter
+                kwargs:
+                max_step: 8
+                data_ticks: 240
+                data_dim: 6
+                processed_data_provider:
+                    class: PickleProcessedDataProvider
+                    module_path: qlib.rl.data.pickle_styled
+                    kwargs:
+                    data_dir: ./data/pickle_dataframe/feature
+            action_interpreter: 
+                class: CategoricalActionInterpreter
+                module_path: qlib.rl.order_execution.interpreter
+                kwargs: 
+                values: 14
+                max_step: 8
+            network: 
+                class: Recurrent
+                module_path: qlib.rl.order_execution.network
+                kwargs: {}
+            policy: 
+                class: PPO
+                module_path: qlib.rl.order_execution.policy
+                kwargs: 
+                    lr: 1.0e-4
+                    # Local path to the latest model. The model is generated during training, so please run training first if you want to run backtest with a trained policy. You could also remove this parameter file to run backtest with a randomly initialized policy.
+                    weight_file: ./checkpoints/latest.pth
+    # Concurrent environment workers.
+    concurrency: 5
+
+With the above config files, you can start training the agent by the following command:
+
+.. code-block:: console
+
+    $ python -m qlib.rl.contrib.train_onpolicy.py --config_path train_config.yml
+
+After the training, you can backtest with the following command:
+
+.. code-block:: console
+
+    $ python -m qlib.rl.contrib.backtest.py --config_path backtest_config.yml
+
+In that case, :class:`~qlib.rl.order_execution.simulator_qlib.SingleAssetOrderExecution` and :class:`~qlib.rl.order_execution.simulator_simple.SingleAssetOrderExecutionSimple` as examples for simulator, :class:`qlib.rl.order_execution.interpreter.FullHistoryStateInterpreter` and :class:`qlib.rl.order_execution.interpreter.CategoricalActionInterpreter` as examples for interpreter, :class:`qlib.rl.order_execution.policy.PPO` as an example for policy, and :class:`qlib.rl.order_execution.reward.PAPenaltyReward` as an example for reward.
+For the single asset order execution task, if developers have already defined their simulator/interpreters/reward function/policy, they could launch the training and backtest pipeline by simply modifying the corresponding settings in the config files.
+The details about the example can be found `here <https://github.com/microsoft/qlib/blob/main/examples/rl/README.md>`_. 
+
+In the future, we will provide more examples for different scenarios such as RL-based portfolio construction.

docs/component/rl/toctree.rst

@@ -0,0 +1,10 @@
+.. _rl:
+
+========================================================================
+Reinforcement Learning in Quantitative Trading
+========================================================================
+
+.. toctree::
+    Overall <overall>
+    Quick Start <quickstart>
+    Framework <framework>

docs/component/strategy.rst

@@ -6,7 +6,7 @@ Portfolio Strategy: Portfolio Management
 .. currentmodule:: qlib
 
 Introduction
-===================
+============
 
 ``Portfolio Strategy`` is designed to adopt different portfolio strategies, which means that users can adopt different algorithms to generate investment portfolios based on the prediction scores of the ``Forecast Model``. Users can use the ``Portfolio Strategy`` in an automatic workflow by ``Workflow`` module, please refer to `Workflow: Workflow Management <workflow.html>`_.
 
@@ -20,7 +20,7 @@ Base Class & Interface
 ======================
 
 BaseStrategy
-------------------
+------------
 
 Qlib provides a base class ``qlib.strategy.base.BaseStrategy``. All strategy classes need to inherit the base class and implement its interface.
 
@@ -32,7 +32,7 @@ Qlib provides a base class ``qlib.strategy.base.BaseStrategy``. All strategy cla
 Users can inherit `BaseStrategy` to customize their strategy class.
 
 WeightStrategyBase
---------------------
+------------------
 
 Qlib also provides a class ``qlib.contrib.strategy.WeightStrategyBase`` that is a subclass of `BaseStrategy`.
 
@@ -60,13 +60,13 @@ Implemented Strategy
 Qlib provides a implemented strategy classes named `TopkDropoutStrategy`.
 
 TopkDropoutStrategy
-------------------
+-------------------
 `TopkDropoutStrategy` is a subclass of `BaseStrategy` and implement the interface `generate_order_list` whose process is as follows.
 
 - Adopt the ``Topk-Drop`` algorithm to calculate the target amount of each stock
 
     .. note::
-        There are two parameters for the ``Topk-Drop`` algorithm：
+        There are two parameters for the ``Topk-Drop`` algorithm:
 
         - `Topk`: The number of stocks held
         - `Drop`: The number of stocks sold on each trading day
@@ -74,16 +74,17 @@ TopkDropoutStrategy
         In general, the number of stocks currently held is `Topk`, with the exception of being zero at the beginning period of trading.
         For each trading day, let $d$ be the number of the instruments currently held and with a rank $\gt K$ when ranked by the prediction scores from high to low.
         Then `d` number of stocks currently held with the worst `prediction score` will be sold, and the same number of unheld stocks with the best `prediction score` will be bought.
-        
+
         In general, $d=$`Drop`, especially when the pool of the candidate instruments is large, $K$ is large, and `Drop` is small.
-         
+
         In most cases, ``TopkDrop`` algorithm sells and buys `Drop` stocks every trading day, which yields a turnover rate of 2$\times$`Drop`/$K$.
-         
+
         The following images illustrate a typical scenario.
+
         .. image:: ../_static/img/topk_drop.png
             :alt: Topk-Drop
 
-       
+
 
 - Generate the order list from the target amount
 
@@ -98,12 +99,12 @@ and `qlib.contrib.strategy.optimizer.enhanced_indexing.EnhancedIndexingOptimizer
 
 
 Usage & Example
-====================
+===============
 
 First, user can create a model to get trading signals(the variable name is ``pred_score`` in following cases).
 
 Prediction Score
------------------
+----------------
 
 The `prediction score` is a pandas DataFrame. Its index is <datetime(pd.Timestamp), instrument(str)> and it must
 contains a `score` column.
@@ -134,7 +135,7 @@ Qlib didn't add a step to scale the prediction score to a unified scale due to t
 - The model has the flexibility to define the target, loss, and data processing. So we don't think there is a silver bullet to rescale it back directly barely based on the model's outputs. If you want to scale it back to some meaningful values(e.g. stock returns.), an intuitive solution is to create a regression model for the model's recent outputs and your recent target values.
 
 Running backtest
------------------
+----------------
 
 - In most cases, users could backtest their portfolio management strategy  with ``backtest_daily``.
 
@@ -195,7 +196,7 @@ Running backtest
 
         CSI300_BENCH = "SH000300"
         # Benchmark is for calculating the excess return of your strategy.
-        # Its data format will be like **ONE normal instrument**. 
+        # Its data format will be like **ONE normal instrument**.
         # For example, you can query its data with the code below
         # `D.features(["SH000300"], ["$close"], start_time='2010-01-01', end_time='2017-12-31', freq='day')`
         # It is different from the argument `market`, which indicates a universe of stocks (e.g. **A SET** of stocks like csi300)
@@ -262,7 +263,7 @@ Running backtest
 
 
 Result
-------------------
+------
 
 The backtest results are in the following form:
 
@@ -307,5 +308,5 @@ The backtest results are in the following form:
 
 
 Reference
-===================
+=========
 To know more about the `prediction score` `pred_score` output by ``Forecast Model``, please refer to `Forecast Model: Model Training & Prediction <model.html>`_.

docs/component/workflow.rst

@@ -1,12 +1,12 @@
 .. _workflow:
 
-=================================
+=============================
 Workflow: Workflow Management
-=================================
+=============================
 .. currentmodule:: qlib
 
 Introduction
-===================
+============
 
 The components in `Qlib Framework <../introduction/introduction.html#framework>`_ are designed in a loosely-coupled way. Users could build their own Quant research workflow with these components like `Example <https://github.com/microsoft/qlib/blob/main/examples/workflow_by_code.py>`_.
 
@@ -28,7 +28,7 @@ With ``qrun``, user can easily start an `execution`, which includes the followin
 For each `execution`, ``Qlib`` has a complete system to tracking all the information as well as artifacts generated during training, inference and evaluation phase. For more information about how ``Qlib`` handles this, please refer to the related document: `Recorder: Experiment Management <../component/recorder.html>`_.
 
 Complete Example
-===================
+================
 
 Before getting into details, here is a complete example of ``qrun``, which defines the workflow in typical Quant research.
 Below is a typical config file of ``qrun``.
@@ -54,7 +54,7 @@ Below is a typical config file of ``qrun``.
                 topk: 50
                 n_drop: 5
                 signal:
-                    - <MODEL> 
+                    - <MODEL>
                     - <DATASET>
         backtest:
             limit_threshold: 0.095
@@ -90,13 +90,13 @@ Below is a typical config file of ``qrun``.
                     train: [2008-01-01, 2014-12-31]
                     valid: [2015-01-01, 2016-12-31]
                     test: [2017-01-01, 2020-08-01]
-        record: 
+        record:
             - class: SignalRecord
               module_path: qlib.workflow.record_temp
               kwargs: {}
             - class: PortAnaRecord
               module_path: qlib.workflow.record_temp
-              kwargs: 
+              kwargs:
                   config: *port_analysis_config
 
 After saving the config into `configuration.yaml`, users could start the workflow and test their ideas with a single command below.
@@ -111,22 +111,22 @@ If users want to use ``qrun`` under debug mode, please use the following command
 
     python -m pdb qlib/workflow/cli.py examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
 
-.. note:: 
+.. note::
 
     `qrun` will be placed in your $PATH directory when installing ``Qlib``.
 
-.. note:: 
-        
+.. note::
+
     The symbol `&` in `yaml` file stands for an anchor of a field, which is useful when another fields include this parameter as part of the value. Taking the configuration file above as an example, users can directly change the value of `market` and `benchmark` without traversing the entire configuration file.
 
 
 Configuration File
-===================
+==================
 
 Let's get into details of ``qrun`` in this section.
 Before using ``qrun``, users need to prepare a configuration file. The following content shows how to prepare each part of the configuration file.
 
-The design logic of the configuration file is very simple. It predefines fixed workflows and provide this yaml interface to users to define how to initialize each component. 
+The design logic of the configuration file is very simple. It predefines fixed workflows and provide this yaml interface to users to define how to initialize each component.
 It follow the design of `init_instance_by_config <https://github.com/microsoft/qlib/blob/2aee9e0145decc3e71def70909639b5e5a6f4b58/qlib/utils/__init__.py#L264>`_ .  It defines the initialization of each component of Qlib, which typically include the class and the initialization arguments.
 
 For example, the following yaml and code are equivalent.
@@ -166,7 +166,7 @@ For example, the following yaml and code are equivalent.
 
 
 Qlib Init Section
---------------------
+-----------------
 
 At first, the configuration file needs to contain several basic parameters which will be used for qlib initialization.
 
@@ -181,21 +181,21 @@ The meaning of each field is as follows:
     Type: str. The URI of the Qlib data. For example, it could be the location where the data loaded by ``get_data.py`` are stored.
 
 - `region`
-    - If `region` == "us", ``Qlib`` will be initialized in US-stock mode. 
+    - If `region` == "us", ``Qlib`` will be initialized in US-stock mode.
     - If `region` == "cn", ``Qlib`` will be initialized in China-stock mode.
 
-    .. note:: 
-        
+    .. note::
+
         The value of `region` should be aligned with the data stored in `provider_uri`.
 
 
 Task Section
---------------------
+------------
 
 The `task` field in the configuration corresponds to a `task`, which contains the parameters of three different subsections: `Model`, `Dataset` and `Record`.
 
 Model Section
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~
 
 In the `task` field, the `model` section describes the parameters of the model to be used for training and inference. For more information about the base ``Model`` class, please refer to `Qlib Model <../component/model.html>`_.
 
@@ -224,14 +224,14 @@ The meaning of each field is as follows:
     Type: str. The path for the model in qlib.
 
 - `kwargs`
-    The keywords arguments for the model. Please refer to the specific model implementation for more information: `models <https://github.com/microsoft/qlib/blob/main/qlib/contrib/model>`_. 
+    The keywords arguments for the model. Please refer to the specific model implementation for more information: `models <https://github.com/microsoft/qlib/blob/main/qlib/contrib/model>`_.
+
+.. note::
 
-.. note:: 
-        
     ``Qlib`` provides a util named: ``init_instance_by_config`` to initialize any class inside ``Qlib`` with the configuration includes the fields: `class`, `module_path` and `kwargs`.
 
 Dataset Section
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~
 
 The `dataset` field describes the parameters for the ``Dataset`` module in ``Qlib`` as well those for the module ``DataHandler``. For more information about the ``Dataset`` module, please refer to `Qlib Data <../component/data.html#dataset>`_.
 
@@ -266,7 +266,7 @@ Here is the configuration for the ``Dataset`` module which will take care of dat
                 test: [2017-01-01, 2020-08-01]
 
 Record Section
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~
 
 The `record` field is about the parameters the ``Record`` module in ``Qlib``. ``Record`` is responsible for tracking training process and results such as `information Coefficient (IC)` and `backtest` in a standard format.
 
@@ -282,7 +282,7 @@ The following script is the configuration of `backtest` and the `strategy` used
                 topk: 50
                 n_drop: 5
                 signal:
-                    - <MODEL> 
+                    - <MODEL>
                     - <DATASET>
         backtest:
             limit_threshold: 0.095
@@ -299,13 +299,13 @@ Here is the configuration details of different `Record Template` such as ``Signa
 
 .. code-block:: YAML
 
-    record: 
+    record:
         - class: SignalRecord
           module_path: qlib.workflow.record_temp
           kwargs: {}
         - class: PortAnaRecord
           module_path: qlib.workflow.record_temp
-          kwargs: 
+          kwargs:
             config: *port_analysis_config
 
 For more information about the ``Record`` module in ``Qlib``, user can refer to the related document: `Record <../component/recorder.html#record-template>`_.

docs/conf.py

@@ -77,7 +77,7 @@ language = "en_US"
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This patterns also effect to html_static_path and html_extra_path
-exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
+exclude_patterns = ["_build", "Thumbs.db", ".DS_Store", "hidden"]
 
 # The name of the Pygments (syntax highlighting) style to use.
 pygments_style = "sphinx"

docs/developer/code_standard_and_dev_guide.rst

@@ -1,21 +1,22 @@
 .. _code_standard:
 
-=================================
+=============
 Code Standard
-=================================
+=============
 
 Docstring
-=================================
+=========
 Please use the `Numpydoc Style <https://stackoverflow.com/a/24385103>`_.
 
 Continuous Integration
-=================================
-Continuous Integration (CI) tools help you stick to the quality standards by running tests every time you push a new commit and reporting the results to a pull request. 
+======================
+Continuous Integration (CI) tools help you stick to the quality standards by running tests every time you push a new commit and reporting the results to a pull request.
 
 When you submit a PR request, you can check whether your code passes the CI tests in the "check" section at the bottom of the web page.
 
 1. Qlib will check the code format with black. The PR will raise error if your code does not align to the standard of Qlib(e.g. a common error is the mixed use of space and tab).
- You can fix the bug by inputing the following code in the command line.
+
+   You can fix the bug by inputting the following code in the command line.
 
 .. code-block:: bash
 
@@ -23,7 +24,7 @@ When you submit a PR request, you can check whether your code passes the CI test
     python -m black . -l 120
 
 
-2. Qlib will check your code style pylint. The checking command is implemented in [github action workflow](https://github.com/microsoft/qlib/blob/0e8b94a552f1c457cfa6cd2c1bb3b87ebb3fb279/.github/workflows/test.yml#L66). 
+2. Qlib will check your code style pylint. The checking command is implemented in [github action workflow](https://github.com/microsoft/qlib/blob/0e8b94a552f1c457cfa6cd2c1bb3b87ebb3fb279/.github/workflows/test.yml#L66).
    Sometime pylint's restrictions are not that reasonable. You can ignore specific errors like this
 
 .. code-block:: python
@@ -32,7 +33,8 @@ When you submit a PR request, you can check whether your code passes the CI test
 
 
 3. Qlib will check your code style flake8. The checking command is implemented in [github action workflow](https://github.com/microsoft/qlib/blob/0e8b94a552f1c457cfa6cd2c1bb3b87ebb3fb279/.github/workflows/test.yml#L73).
- You can fix the bug by inputing the following code in the command line.
+
+   You can fix the bug by inputing the following code in the command line.
 
 .. code-block:: bash
 
@@ -40,9 +42,22 @@ When you submit a PR request, you can check whether your code passes the CI test
 
 
 4. Qlib has integrated pre-commit, which will make it easier for developers to format their code.
- Just run the following two commands, and the code will be automatically formatted using black and flake8 when the git commit command is executed.
+
+   Just run the following two commands, and the code will be automatically formatted using black and flake8 when the git commit command is executed.
+
+.. code-block:: bash
+
+    pip install -e .[dev]
+    pre-commit install
+
+
+=================================
+Development Guidance
+=================================
+
+As a developer, you often want make changes to `Qlib` and hope it would reflect directly in your environment without reinstalling it. You can install `Qlib` in editable mode with following command.
+The `[dev]` option will help you to install some related packages when developing `Qlib` (e.g. pytest, sphinx)
 
 .. code-block:: bash
 
     pip install -e .[dev]
-    pre-commit install

docs/hidden/client.rst

@@ -1,12 +1,12 @@
 .. _client:
 
 Qlib Client-Server Framework
-===================
+============================
 
 .. currentmodule:: qlib
 
 Introduction
------------
+------------
 Client-Server is designed to solve following  problems
 
 - Manage the data in a centralized way. Users don't have to manage data of different versions.
@@ -81,6 +81,7 @@ If running on Windows, open **NFS** features and write correct **mount_path**, i
     * Open ``Programs and Features``.
     * Click ``Turn Windows features on or off``.
     * Scroll down and check the option ``Services for NFS``, then click OK
+
     Reference address: https://graspingtech.com/mount-nfs-share-windows-10/
 2.config correct mount_path
     * In windows, mount path must be not exist path and root path,
@@ -159,13 +160,11 @@ Limitations
 2. The rolling operation expression with parameter `0` can not be updated rightly under mechanism of the client-server framework.
 
 API
-********************
+***
 
-The client is based on `python-socketio<https://python-socketio.readthedocs.io>`_ which is a framework that supports WebSocket client for Python language. The client can only propose requests and receive results, which do not include any calculating procedure.
+The client is based on `python-socketio <https://python-socketio.readthedocs.io>`_ which is a framework that supports WebSocket client for Python language. The client can only propose requests and receive results, which do not include any calculating procedure.
 
 Class
---------------------
+-----
 
 .. automodule:: qlib.data.client
-
-

docs/hidden/online.rst

@@ -1,11 +1,11 @@
 .. _online:
 
 Online
-===================
+======
 .. currentmodule:: qlib
 
 Introduction
--------------------
+------------
 
 Welcome to use Online, this module simulates what will be like if we do the real trading use our model and strategy.
 
@@ -31,11 +31,11 @@ The file structure can be viewed at fileStruct_.
 
 
 Example
--------------------
+-------
 
 Let's take an example,
 
-.. note:: Make sure you have the latest version of `qlib` installed. 
+.. note:: Make sure you have the latest version of `qlib` installed.
 
 If you want to use the models and data provided by `qlib`, you only need to do as follows.
 
@@ -93,7 +93,7 @@ If Your account was saved in "./user_data/", you can see the performance of your
 Here 'SH000905' represents csi500 and 'SH000300' represents csi300
 
 Manage your account
---------------------
+-------------------
 
 Any account processed by `online` should be saved in a folder. you can use commands
 defined to manage your accounts.
@@ -161,7 +161,7 @@ be called at each trading date.
         >> online update -date 2019-10-16 -path ./user_data/
 
 API
-------------------
+---
 
 All those operations are based on defined in `qlib.contrib.online.operator`
 
@@ -170,7 +170,7 @@ All those operations are based on defined in `qlib.contrib.online.operator`
 .. _fileStruct:
 
 File structure
-------------------
+--------------
 
 'user_data' indicates the root of folder.
 Name that bold indicates it’s a folder, otherwise it’s a document.
@@ -214,7 +214,7 @@ Configuration file
 The configure file used in `online` should contain the model and strategy information.
 
 About the model
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~
 
 First, your configuration file needs to have a field about the model,
 this field and its contents determine the model we used when generating score at predict date.
@@ -243,7 +243,7 @@ contains 2 methods used in `online` module.
 
 
 About the strategy
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~
 
 Your need define the strategy used to generate the order list at predict date.
 
@@ -259,7 +259,7 @@ Followings are two examples for a TopkAmountStrategy
             n_drop: 10
 
 Generated files
-------------------
+---------------
 
 The 'online_generate' command will create the order list at {folder_path}/{user_id}/temp/,
 the name of that is orderlist_{YYYY-MM-DD}.json, YYYY-MM-DD is the date that those orders to be executed.

docs/hidden/tuner.rst

@@ -1,11 +1,11 @@
 .. _tuner:
 
 Tuner
-===================
+=====
 .. currentmodule:: qlib
 
 Introduction
--------------------
+------------
 
 Welcome to use Tuner, this document is based on that you can use Estimator proficiently and correctly.
 
@@ -41,19 +41,19 @@ We write a simple configuration example as following,
         tuner_class: QLibTuner
     qlib_client:
         auto_mount: False
-        logging_level: INFO 
+        logging_level: INFO
     optimization_criteria:
         report_type: model
         report_factor: model_score
         optim_type: max
     tuner_pipeline:
-      - 
-        model: 
+      -
+        model:
             class: SomeModel
             space: SomeModelSpace
-        trainer: 
+        trainer:
             class: RollingTrainer
-        strategy: 
+        strategy:
             class: TopkAmountStrategy
             space: TopkAmountStrategySpace
         max_evals: 2
@@ -166,13 +166,13 @@ Also, there are some optional fields. The meaning of each field is as follows:
     The class of tuner, str type, must be an already implemented model, such as `QLibTuner` in `qlib`, or a custom tuner, but it must be a subclass of `qlib.contrib.tuner.Tuner`, the default value is `QLibTuner`.
 
 - `tuner_module_path`
-    The module path, str type, absolute url is also supported, indicates the path of the implementation of tuner. The default value is `qlib.contrib.tuner.tuner` 
+    The module path, str type, absolute url is also supported, indicates the path of the implementation of tuner. The default value is `qlib.contrib.tuner.tuner`
 
 About the optimization criteria
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 You need to designate a factor to optimize, for tuner need a factor to decide which case is better than other cases.
-Usually, we use the result of `estimator`, such as backtest results and the score of model. 
+Usually, we use the result of `estimator`, such as backtest results and the score of model.
 
 This part needs contain these fields:
 
@@ -203,13 +203,13 @@ The tuner pipeline contains different tuners, and the `tuner` program will proce
 .. code-block:: YAML
 
     tuner_pipeline:
-      - 
-        model: 
+      -
+        model:
             class: SomeModel
             space: SomeModelSpace
-        trainer: 
+        trainer:
             class: RollingTrainer
-        strategy: 
+        strategy:
             class: TopkAmountStrategy
             space: TopkAmountStrategySpace
         max_evals: 2
@@ -249,25 +249,25 @@ You need to use the same dataset to evaluate your different `estimator` experime
         test_start_date: 2016-07-01
         test_end_date: 2018-04-30
 
-- `rolling_period`    
+- `rolling_period`
     The rolling period, integer type, indicates how many time steps need rolling when rolling the data. The default value is `60`. If you use `RollingTrainer`, this config will be used, or it will be ignored.
 
 - `train_start_date`
     Training start time, str type.
 
-- `train_end_date`      
+- `train_end_date`
     Training end time, str type.
 
-- `validate_start_date`    
+- `validate_start_date`
     Validation start time, str type.
 
-- `validate_end_date`    
+- `validate_end_date`
     Validation end time, str type.
 
-- `test_start_date`    
+- `test_start_date`
     Test start time, str type.
 
-- `test_end_date`     
+- `test_end_date`
     Test end time, str type. If `test_end_date` is `-1` or greater than the last date of the data, the last date of the data will be used as `test_end_date`.
 
 About the data and backtest
@@ -315,11 +315,10 @@ About the data and backtest
 Experiment Result
 -----------------
 
-All the results are stored in experiment file directly, you can check them directly in the corresponding files. 
+All the results are stored in experiment file directly, you can check them directly in the corresponding files.
 What we save are as following:
 
 - Global optimal parameters
 - Local optimal parameters of each tuner
 - Config file of this `tuner` experiment
 - Every `estimator` experiments result in the process
-

docs/index.rst

@@ -1,6 +1,6 @@
-============================================================
+======================
 ``Qlib`` Documentation
-============================================================
+======================
 
 ``Qlib`` is an AI-oriented quantitative investment platform, which aims to realize the potential, empower the research, and create the value of AI technologies in quantitative investment.
 
@@ -24,16 +24,16 @@ Document Structure
 .. toctree::
    :maxdepth: 3
    :caption: FIRST STEPS:
-   
+
    Installation <start/installation.rst>
    Initialization <start/initialization.rst>
    Data Retrieval <start/getdata.rst>
    Custom Model Integration <start/integration.rst>
-   
+
 
 .. toctree::
    :maxdepth: 3
-   :caption: COMPONENTS:
+   :caption: MAIN COMPONENTS:
 
    Workflow: Workflow Management <component/workflow.rst>
    Data Layer: Data Framework & Usage <component/data.rst>
@@ -44,17 +44,24 @@ Document Structure
    Qlib Recorder: Experiment Management <component/recorder.rst>
    Analysis: Evaluation & Results Analysis <component/report.rst>
    Online Serving: Online Management & Strategy & Tool <component/online.rst>
+   Reinforcement Learning <component/rl/toctree>
 
 .. toctree::
    :maxdepth: 3
-   :caption: ADVANCED TOPICS:
-   
+   :caption: OTHER COMPONENTS/FEATURES/TOPICS:
+
    Building Formulaic Alphas <advanced/alpha.rst>
    Online & Offline mode <advanced/server.rst>
    Serialization <advanced/serial.rst>
    Task Management <advanced/task_management.rst>
    Point-In-Time database <advanced/PIT.rst>
 
+.. toctree::
+   :maxdepth: 3
+   :caption: FOR DEVELOPERS:
+
+   Code Standard & Development Guidance <developer/code_standard_and_dev_guide.rst>
+
 .. toctree::
    :maxdepth: 3
    :caption: REFERENCE:

docs/introduction/introduction.rst

@@ -3,7 +3,7 @@
 ===============================
 
 Introduction
-===================
+============
 
 .. image:: ../_static/img/logo/white_bg_rec+word.png
     :align: center
@@ -13,40 +13,58 @@ Introduction
 With ``Qlib``, users can easily try their ideas to create better Quant investment strategies.
 
 Framework
-===================
-   
+=========
+
+
 .. image:: ../_static/img/framework.svg
     :align: center
 
 
 At the module level, Qlib is a platform that consists of above components. The components are designed as loose-coupled modules and each component could be used stand-alone.
 
+This framework may be intimidating for new users to Qlib. It tries to accurately include a lot of details of Qlib's design.
+For users new to Qlib, you can skip it first and read it later.
 
 
-========================  ==============================================================================
-Name                      Description
-========================  ==============================================================================
-`Infrastructure` layer    `Infrastructure` layer provides underlying support for Quant research.
-                          `DataServer` provides high-performance infrastructure for users to manage 
-                          and retrieve raw data. `Trainer` provides flexible interface to control
-                          the training process of models which enable algorithms controlling the
-                          training process.
 
-`Workflow` layer          `Workflow` layer covers the whole workflow of quantitative investment.
-                          `Information Extractor` extracts data for models. `Forecast Model` focuses
-                          on producing all kinds of forecast signals (e.g. *alpha*, risk) for other
-                          modules.  With these signals `Decision Generator` will generate the target 
-                          trading decisions(i.e. portfolio, orders)  to be executed by `Execution Env`
-                          (i.e. the trading market).  There may be multiple levels of `Trading Agent`
-                          and `Execution Env` (e.g. an *order executor trading agent and intraday
-                          order execution environment* could behave like an interday trading
-                          environment and nested in  *daily portfolio management trading agent and
-                          interday trading environment*  ) 
+===========================  ==============================================================================
+Name                         Description
+===========================  ==============================================================================
+`Infrastructure` layer       `Infrastructure` layer provides underlying support for Quant research.
+                             `DataServer` provides high-performance infrastructure for users to manage
+                             and retrieve raw data. `Trainer` provides flexible interface to control
+                             the training process of models which enable algorithms controlling the
+                             training process.
 
-`Interface` layer         `Interface` layer tries to present a user-friendly interface for the underlying
-                          system. `Analyser` module will provide users detailed analysis reports of
-                          forecasting signals, portfolios and execution results
-========================  ==============================================================================
+`Learning Framework` layer   The `Forecast Model` and `Trading Agent` are learnable. They are learned
+                             based on the `Learning Framework` layer and then applied to multiple scenarios
+                             in `Workflow` layer. The supported learning paradigms can be categorized into
+                             reinforcement learning and supervised learning.  The learning framework
+                             leverages the `Workflow` layer as well(e.g. sharing `Information Extractor`,
+                             creating environments based on `Execution Env`).
+
+`Workflow` layer             `Workflow` layer covers the whole workflow of quantitative investment.
+                             Both supervised-learning-based strategies and RL-based Strategies
+                             are supported.
+                             `Information Extractor` extracts data for models. `Forecast Model` focuses
+                             on producing all kinds of forecast signals (e.g. *alpha*, risk) for other
+                             modules.  With these signals `Decision Generator` will generate the target
+                             trading decisions(i.e. portfolio, orders)
+                             If RL-based Strategies are adopted, the `Policy` is learned in a end-to-end way,
+                             the trading deicsions are generated directly.
+                             Decisions will be executed by `Execution Env`
+                             (i.e. the trading market).  There may be multiple levels of `Strategy`
+                             and `Executor` (e.g. an *order executor trading strategy and intraday order executor*
+                             could behave like an interday trading loop and be nested in
+                             *daily portfolio management trading strategy and interday trading executor*
+                             trading loop)
+
+`Interface` layer            `Interface` layer tries to present a user-friendly interface for the underlying
+                             system. `Analyser` module will provide users detailed analysis reports of
+                             forecasting signals, portfolios and execution results
+===========================  ==============================================================================
 
 - The modules with hand-drawn style are under development and will be released in the future.
 - The modules with dashed borders are highly user-customizable and extendible.
+
+(p.s. framework image is created with https://draw.io/)

docs/introduction/quick.rst

@@ -1,10 +1,10 @@
 
-===============================
+===========
 Quick Start
-===============================
+===========
 
 Introduction
-==============
+============
 
 This ``Quick Start`` guide tries to demonstrate
 
@@ -14,13 +14,14 @@ This ``Quick Start`` guide tries to demonstrate
 
 
 Installation
-==================
+============
 
 Users can easily intsall ``Qlib`` according to the following steps:
 
 - Before installing ``Qlib`` from source, users need to install some dependencies:
 
     .. code-block::
+
         pip install numpy
         pip install --upgrade  cython
 
@@ -34,7 +35,7 @@ Users can easily intsall ``Qlib`` according to the following steps:
 To known more about `installation`, please refer to `Qlib Installation <../start/installation.html>`_.
 
 Prepare Data
-==============
+============
 
 Load and prepare data by running the following code:
 
@@ -47,14 +48,14 @@ This dataset is created by public data collected by crawler scripts in ``scripts
 To known more about `prepare data`, please refer to `Data Preparation <../component/data.html#data-preparation>`_.
 
 Auto Quant Research Workflow
-====================================
+============================
 
-``Qlib`` provides a tool named ``qrun`` to run the whole workflow automatically (including building dataset, training models, backtest and evaluation). Users can start an auto quant research workflow and have a graphical reports analysis according to the following steps: 
+``Qlib`` provides a tool named ``qrun`` to run the whole workflow automatically (including building dataset, training models, backtest and evaluation). Users can start an auto quant research workflow and have a graphical reports analysis according to the following steps:
 
-- Quant Research Workflow: 
+- Quant Research Workflow:
     - Run  ``qrun`` with a config file of the LightGBM model `workflow_config_lightgbm.yaml` as following.
 
-        .. code-block:: 
+        .. code-block::
 
             cd examples  # Avoid running program under the directory contains `qlib`
             qrun benchmarks/LightGBM/workflow_config_lightgbm.yaml
@@ -64,7 +65,7 @@ Auto Quant Research Workflow
         The result of ``qrun`` is as follows, which is also the typical result of ``Forecast model(alpha)``. Please refer to  `Intraday Trading <../component/backtest.html>`_. for more details about the result.
 
         .. code-block:: python
-        
+
                                                               risk
             excess_return_without_cost mean               0.000605
                                        std                0.005481
@@ -77,7 +78,7 @@ Auto Quant Research Workflow
                                        information_ratio  1.187411
                                        max_drawdown      -0.075024
 
-        
+
     To know more about `workflow` and `qrun`, please refer to `Workflow: Workflow Management <../component/workflow.html>`_.
 
 - Graphical Reports Analysis:
@@ -89,6 +90,6 @@ Auto Quant Research Workflow
 
 
 Custom Model Integration
-===============================================
+========================
 
 ``Qlib`` provides a batch of models (such as ``lightGBM`` and ``MLP`` models) as examples of ``Forecast Model``. In addition to the default model, users can integrate their own custom models into ``Qlib``. If users are interested in the custom model, please refer to `Custom Model Integration <../start/integration.html>`_.

docs/make.bat

@@ -0,0 +1,35 @@
+@ECHO OFF
+
+pushd %~dp0
+
+REM Command file for Sphinx documentation
+
+if "%SPHINXBUILD%" == "" (
+	set SPHINXBUILD=sphinx-build
+)
+set SOURCEDIR=.
+set BUILDDIR=_build
+
+%SPHINXBUILD% >NUL 2>NUL
+if errorlevel 9009 (
+	echo.
+	echo.The 'sphinx-build' command was not found. Make sure you have Sphinx
+	echo.installed, then set the SPHINXBUILD environment variable to point
+	echo.to the full path of the 'sphinx-build' executable. Alternatively you
+	echo.may add the Sphinx directory to PATH.
+	echo.
+	echo.If you don't have Sphinx installed, grab it from
+	echo.https://www.sphinx-doc.org/
+	exit /b 1
+)
+
+if "%1" == "" goto help
+
+%SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+goto end
+
+:help
+%SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
+
+:end
+popd

docs/reference/api.rst

@@ -1,7 +1,8 @@
 .. _api:
-================================
+
+=============
 API Reference
-================================
+=============
 
 
 
@@ -9,32 +10,32 @@ Here you can find all ``Qlib`` interfaces.
 
 
 Data
-====================
+====
 
 Provider
---------------------
+--------
 
 .. automodule:: qlib.data.data
     :members:
-		
+
 Filter
---------------------
+------
 
 .. automodule:: qlib.data.filter
     :members:
 
 Class
---------------------
+-----
 .. automodule:: qlib.data.base
     :members:
 
 Operator
---------------------
+--------
 .. automodule:: qlib.data.ops
     :members:
-	       
+
 Cache
-----------------
+-----
 .. autoclass:: qlib.data.cache.MemCacheUnit
     :members:
 
@@ -55,7 +56,7 @@ Cache
 
 
 Storage
--------------
+-------
 .. autoclass:: qlib.data.storage.storage.BaseStorage
     :members:
 
@@ -82,52 +83,52 @@ Storage
 
 
 Dataset
----------------
+-------
 
 Dataset Class
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~
 .. automodule:: qlib.data.dataset.__init__
     :members:
 
 Data Loader
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~
 .. automodule:: qlib.data.dataset.loader
     :members:
 
 Data Handler
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~
 .. automodule:: qlib.data.dataset.handler
     :members:
 
 Processor
-~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~
 .. automodule:: qlib.data.dataset.processor
     :members:
 
 
 Contrib
-====================
+=======
 
 Model
---------------------
+-----
 .. automodule:: qlib.model.base
     :members:
 
 Strategy
--------------------
+--------
 
-.. automodule:: qlib.contrib.strategy.strategy
+.. automodule:: qlib.contrib.strategy
     :members:
 
 Evaluate
------------------
+--------
 
 .. automodule:: qlib.contrib.evaluate
     :members:
-    
+
 
 Report
------------------
+------
 
 .. automodule:: qlib.contrib.report.analysis_position.report
     :members:
@@ -159,103 +160,133 @@ Report
 
 
 Workflow
-====================
+========
 
 
 Experiment Manager
---------------------
+------------------
 .. autoclass:: qlib.workflow.expm.ExpManager
     :members:
 
 Experiment
---------------------
+----------
 .. autoclass:: qlib.workflow.exp.Experiment
     :members:
 
 Recorder
---------------------
+--------
 .. autoclass:: qlib.workflow.recorder.Recorder
     :members:
 
 Record Template
---------------------
+---------------
 .. automodule:: qlib.workflow.record_temp
     :members:
 
 Task Management
-====================
+===============
 
 
 TaskGen
---------------------
+-------
 .. automodule:: qlib.workflow.task.gen
     :members:
 
 TaskManager
---------------------
+-----------
 .. automodule:: qlib.workflow.task.manage
     :members:
 
 Trainer
---------------------
+-------
 .. automodule:: qlib.model.trainer
     :members:
 
 Collector
---------------------
+---------
 .. automodule:: qlib.workflow.task.collect
     :members:
 
 Group
---------------------
+-----
 .. automodule:: qlib.model.ens.group
     :members:
 
 Ensemble
---------------------
+--------
 .. automodule:: qlib.model.ens.ensemble
     :members:
 
 Utils
---------------------
+-----
 .. automodule:: qlib.workflow.task.utils
     :members:
 
 
 Online Serving
-====================
+==============
 
 
 Online Manager
---------------------
+--------------
 .. automodule:: qlib.workflow.online.manager
     :members:
 
 Online Strategy
---------------------
+---------------
 .. automodule:: qlib.workflow.online.strategy
     :members:
 
 Online Tool
---------------------
+-----------
 .. automodule:: qlib.workflow.online.utils
     :members:
 
 
 RecordUpdater
---------------------
+-------------
 .. automodule:: qlib.workflow.online.update
     :members:
 
 
 Utils
-====================
+=====
 
 Serializable
---------------------
+------------
 
-.. automodule:: qlib.utils.serial.Serializable
+.. automodule:: qlib.utils.serial
     :members:
 
+RL
+==============
 
-    
+Base Component
+--------------
+.. automodule:: qlib.rl
+    :members:
+    :imported-members:
+
+Strategy
+--------
+.. automodule:: qlib.rl.strategy
+    :members:
+    :imported-members:
+
+Trainer
+-------
+.. automodule:: qlib.rl.trainer
+    :members:
+    :imported-members:
+
+Order Execution
+---------------
+.. automodule:: qlib.rl.order_execution
+    :members:
+    :imported-members:
+
+Utils
+---------------
+.. automodule:: qlib.rl.utils
+    :members:
+    :imported-members:

docs/requirements.txt

@@ -4,3 +4,4 @@ numpy
 scipy
 scikit-learn
 pandas
+tianshou

docs/start/getdata.rst

@@ -1,18 +1,18 @@
 .. _getdata:
 
-=============================
+==============
 Data Retrieval
-=============================
+==============
 
 .. currentmodule:: qlib
 
 Introduction
-====================
+============
 
 Users can get stock data with ``Qlib``. The following examples demonstrate the basic user interface.
 
 Examples
-====================
+========
 
 
 ``QLib`` Initialization:
@@ -30,7 +30,7 @@ If users followed steps in `initialization <initialization.html>`_ and downloade
 Load trading calendar with given time range and frequency:
 
 .. code-block:: python
-		
+
    >> from qlib.data import D
    >> D.calendar(start_time='2010-01-01', end_time='2017-12-31', freq='day')[:2]
    [Timestamp('2010-01-04 00:00:00'), Timestamp('2010-01-05 00:00:00')]
@@ -46,7 +46,7 @@ Parse a given market name into a stock pool config:
 Load instruments of certain stock pool in the given time range:
 
 .. code-block:: python
-		
+
    >> from qlib.data import D
    >> instruments = D.instruments(market='csi300')
    >> D.list_instruments(instruments=instruments, start_time='2010-01-01', end_time='2017-12-31', as_list=True)[:6]
@@ -79,19 +79,18 @@ For more details about filter, please refer `Filter API <../component/data.html>
 Load features of certain instruments in a given time range:
 
 .. code-block:: python
-		
+
    >> from qlib.data import D
    >> instruments = ['SH600000']
    >> fields = ['$close', '$volume', 'Ref($close, 1)', 'Mean($close, 3)', '$high-$low']
-   >> D.features(instruments, fields, start_time='2010-01-01', end_time='2017-12-31', freq='day').head()
-                              
-                              $close     $volume  Ref($close, 1)  Mean($close, 3)  $high-$low
-      instrument  datetime                                                                      
-      SH600000    2010-01-04  86.778313  16162960.0       88.825928        88.061483    2.907631
-                  2010-01-05  87.433578  28117442.0       86.778313        87.679273    3.235252
-                  2010-01-06  85.713585  23632884.0       87.433578        86.641825    1.720009
-                  2010-01-07  83.788803  20813402.0       85.713585        85.645322    3.030487
-                  2010-01-08  84.730675  16044853.0       83.788803        84.744354    2.047623
+   >> D.features(instruments, fields, start_time='2010-01-01', end_time='2017-12-31', freq='day').head().to_string()
+   '                           $close     $volume  Ref($close, 1)  Mean($close, 3)  $high-$low
+   ... instrument  datetime
+   ... SH600000    2010-01-04  86.778313  16162960.0       88.825928        88.061483    2.907631
+   ...             2010-01-05  87.433578  28117442.0       86.778313        87.679273    3.235252
+   ...             2010-01-06  85.713585  23632884.0       87.433578        86.641825    1.720009
+   ...             2010-01-07  83.788803  20813402.0       85.713585        85.645322    3.030487
+   ...             2010-01-08  84.730675  16044853.0       83.788803        84.744354    2.047623'
 
 Load features of certain stock pool in a given time range:
 
@@ -105,15 +104,14 @@ Load features of certain stock pool in a given time range:
    >> expressionDFilter = ExpressionDFilter(rule_expression='$close>Ref($close,1)')
    >> instruments = D.instruments(market='csi300', filter_pipe=[nameDFilter, expressionDFilter])
    >> fields = ['$close', '$volume', 'Ref($close, 1)', 'Mean($close, 3)', '$high-$low']
-   >> D.features(instruments, fields, start_time='2010-01-01', end_time='2017-12-31', freq='day').head()
-
-                                 $close        $volume  Ref($close, 1)  Mean($close, 3)  $high-$low
-      instrument  datetime                                                                           
-      SH600655    2010-01-04  2699.567383  158193.328125     2619.070312      2626.097738  124.580566
-                  2010-01-08  2612.359619   77501.406250     2584.567627      2623.220133   83.373047
-                  2010-01-11  2712.982422  160852.390625     2612.359619      2636.636556  146.621582
-                  2010-01-12  2788.688232  164587.937500     2712.982422      2704.676758  128.413818
-                  2010-01-13  2790.604004  145460.453125     2788.688232      2764.091553  128.413818
+   >> D.features(instruments, fields, start_time='2010-01-01', end_time='2017-12-31', freq='day').head().to_string()
+   '                              $close        $volume  Ref($close, 1)  Mean($close, 3)  $high-$low
+   ... instrument  datetime
+   ... SH600655    2010-01-04  2699.567383  158193.328125     2619.070312      2626.097738  124.580566
+   ...             2010-01-08  2612.359619   77501.406250     2584.567627      2623.220133   83.373047
+   ...             2010-01-11  2712.982422  160852.390625     2612.359619      2636.636556  146.621582
+   ...             2010-01-12  2788.688232  164587.937500     2712.982422      2704.676758  128.413818
+   ...             2010-01-13  2790.604004  145460.453125     2788.688232      2764.091553  128.413818'
 
 
 For more details about features, please refer `Feature API <../component/data.html>`_.
@@ -127,7 +125,7 @@ For example, it looks quite long and complicated:
 .. code-block:: python
 
    >> from qlib.data import D
-   >> data = D.features(["sh600519"], ["(($high / $close) + ($open / $close)) * (($high / $close) + ($open / $close)) / ($high / $close) + ($open / $close)"], start_time="20200101")
+   >> data = D.features(["sh600519"], ["(($high / $close) + ($open / $close)) * (($high / $close) + ($open / $close)) / (($high / $close) + ($open / $close))"], start_time="20200101")
 
 
 But using string is not the only way to implement the expression. You can also implement expression by code.
@@ -147,5 +145,5 @@ Here is an exmaple which does the same thing as above examples.
 
 
 API
-====================
+===
 To know more about how to use the Data, go to API Reference: `Data API <../reference/api.html#data>`_

docs/start/initialization.rst

@@ -1,23 +1,23 @@
 .. _initialization:
 
-====================
+===================
 Qlib Initialization
-====================
+===================
 
 .. currentmodule:: qlib
 
 
 Initialization
-=========================
+==============
 
 Please follow the steps below to initialize ``Qlib``.
 
 Download and prepare the Data: execute the following command to download stock data. Please pay `attention` that the data is collected from `Yahoo Finance <https://finance.yahoo.com/lookup>`_ and the data might not be perfect. We recommend users to prepare their own data if they have high-quality datasets. Please refer to `Data <../component/data.html#converting-csv-format-into-qlib-format>`_ for more information about customized dataset.
-    
+
     .. code-block:: bash
-    
+
         python scripts/get_data.py qlib_data --target_dir ~/.qlib/qlib_data/cn_data --region cn
-        
+
 Please refer to `Data Preparation <../component/data.html#data-preparation>`_ for more information about `get_data.py`,
 
 
@@ -30,7 +30,7 @@ Initialize Qlib before calling other APIs: run following code in python.
         from qlib.constant import REG_CN
         provider_uri = "~/.qlib/qlib_data/cn_data"  # target_dir
         qlib.init(provider_uri=provider_uri, region=REG_CN)
-    
+
 .. note::
    Do not import qlib package in the repository directory  of ``Qlib``, otherwise, errors may occur.
 
@@ -56,16 +56,16 @@ The following are several important parameters of `qlib.init` (`Qlib` has a lot
 - `redis_port`
     Type: int, optional parameter(default: 6379), port of `redis`
 
-    .. note:: 
-        
+    .. note::
+
         The value of `region` should be aligned with the data stored in `provider_uri`. Currently, ``scripts/get_data.py`` only provides China stock market data. If users want to use the US stock market data, they should prepare their own US-stock data in `provider_uri` and switch to US-stock mode.
 
     .. note::
-        
+
         If Qlib fails to connect redis via `redis_host` and `redis_port`, cache mechanism will not be used! Please refer to `Cache <../component/data.html#cache>`_ for details.
 - `exp_manager`
     Type: dict, optional parameter, the setting of `experiment manager` to be used in qlib. Users can specify an experiment manager class, as well as the tracking URI for all the experiments. However, please be aware that we only support input of a dictionary in the following style for `exp_manager`. For more information about `exp_manager`, users can refer to `Recorder: Experiment Management <../component/recorder.html>`_.
-    
+
     .. code-block:: Python
 
         # For example, if you want to set your tracking_uri to a <specific folder>, you can initialize qlib below
@@ -78,7 +78,7 @@ The following are several important parameters of `qlib.init` (`Qlib` has a lot
             }
         })
 - `mongo`
-    Type: dict, optional parameter, the setting of `MongoDB <https://www.mongodb.com/>`_ which will be used in some features such as `Task Management <../advanced/task_management.html>`_, with high performance and clustered processing. 
+    Type: dict, optional parameter, the setting of `MongoDB <https://www.mongodb.com/>`_ which will be used in some features such as `Task Management <../advanced/task_management.html>`_, with high performance and clustered processing.
     Users need to follow the steps in  `installation <https://www.mongodb.com/try/download/community>`_  to install MongoDB firstly and then access it via a URI.
     Users can access mongodb with credential by setting "task_url"  to a string like `"mongodb://%s:%s@%s" % (user, pwd, host + ":" + port)`.
 

docs/start/installation.rst

@@ -1,8 +1,8 @@
 .. _installation:
 
-====================
+============
 Installation
-====================
+============
 
 .. currentmodule:: qlib
 
@@ -24,7 +24,7 @@ Also, Users can install ``Qlib`` by the source code according to the following s
 
 - Enter the root directory of ``Qlib``, in which the file ``setup.py`` exists.
 - Then, please execute the following command to install the environment dependencies and install ``Qlib``:
-   
+
    .. code-block:: bash
 
       $ pip install numpy
@@ -34,7 +34,7 @@ Also, Users can install ``Qlib`` by the source code according to the following s
 
 .. note::
    It's recommended to use anaconda/miniconda to setup the environment. ``Qlib`` needs lightgbm and pytorch packages, use pip to install them.
-   
+
 
 
 Use the following code to make sure the installation successful:
@@ -44,6 +44,3 @@ Use the following code to make sure the installation successful:
    >>> import qlib
    >>> qlib.__version__
    <LATEST VERSION>
-
-
-=====================

docs/start/integration.rst

@@ -1,9 +1,9 @@
-=========================================
+========================
 Custom Model Integration
-=========================================
+========================
 
 Introduction
-===================
+============
 
 ``Qlib``'s `Model Zoo` includes models such as ``LightGBM``, ``MLP``, ``LSTM``, etc.. These models are examples of ``Forecast Model``. In addition to the default models ``Qlib`` provide, users can integrate their own custom models into ``Qlib``.
 
@@ -14,119 +14,123 @@ Users can integrate their own custom models according to the following steps.
 - Test the custom model.
 
 Custom Model Class
-===========================
+==================
 The Custom models need to inherit `qlib.model.base.Model <../reference/api.html#module-qlib.model.base>`_ and override the methods in it.
 
 - Override the `__init__` method
     - ``Qlib`` passes the initialized parameters to the \_\_init\_\_ method.
     - The hyperparameters of model in the configuration must be consistent with those defined in the `__init__` method.
     - Code Example: In the following example, the hyperparameters of model in the configuration file should contain parameters such as `loss:mse`.
-    .. code-block:: Python
 
-        def __init__(self, loss='mse', **kwargs):
-            if loss not in {'mse', 'binary'}:
-                raise NotImplementedError
-            self._scorer = mean_squared_error if loss == 'mse' else roc_auc_score
-            self._params.update(objective=loss, **kwargs)
-            self._model = None
+        .. code-block:: Python
+
+            def __init__(self, loss='mse', **kwargs):
+                if loss not in {'mse', 'binary'}:
+                    raise NotImplementedError
+                self._scorer = mean_squared_error if loss == 'mse' else roc_auc_score
+                self._params.update(objective=loss, **kwargs)
+                self._model = None
 
 - Override the `fit` method
     - ``Qlib`` calls the fit method to train the model.
     - The parameters must include training feature `dataset`, which is designed in the interface.
     - The parameters could include some `optional` parameters with default values, such as `num_boost_round = 1000` for `GBDT`.
     - Code Example: In the following example, `num_boost_round = 1000` is an optional parameter.
-    .. code-block:: Python
-    
-        def fit(self, dataset: DatasetH, num_boost_round = 1000, **kwargs):
 
-            # prepare dataset for lgb training and evaluation
-            df_train, df_valid = dataset.prepare(
-                ["train", "valid"], col_set=["feature", "label"], data_key=DataHandlerLP.DK_L
-            )
-            x_train, y_train = df_train["feature"], df_train["label"]
-            x_valid, y_valid = df_valid["feature"], df_valid["label"]
+        .. code-block:: Python
 
-            # Lightgbm need 1D array as its label
-            if y_train.values.ndim == 2 and y_train.values.shape[1] == 1:
-                y_train, y_valid = np.squeeze(y_train.values), np.squeeze(y_valid.values)
-            else:
-                raise ValueError("LightGBM doesn't support multi-label training")
+            def fit(self, dataset: DatasetH, num_boost_round = 1000, **kwargs):
 
-            dtrain = lgb.Dataset(x_train.values, label=y_train)
-            dvalid = lgb.Dataset(x_valid.values, label=y_valid)
+                # prepare dataset for lgb training and evaluation
+                df_train, df_valid = dataset.prepare(
+                    ["train", "valid"], col_set=["feature", "label"], data_key=DataHandlerLP.DK_L
+                )
+                x_train, y_train = df_train["feature"], df_train["label"]
+                x_valid, y_valid = df_valid["feature"], df_valid["label"]
 
-            # fit the model
-            self.model = lgb.train(
-                self.params,
-                dtrain,
-                num_boost_round=num_boost_round,
-                valid_sets=[dtrain, dvalid],
-                valid_names=["train", "valid"],
-                early_stopping_rounds=early_stopping_rounds,
-                verbose_eval=verbose_eval,
-                evals_result=evals_result,
-                **kwargs
-            )
+                # Lightgbm need 1D array as its label
+                if y_train.values.ndim == 2 and y_train.values.shape[1] == 1:
+                    y_train, y_valid = np.squeeze(y_train.values), np.squeeze(y_valid.values)
+                else:
+                    raise ValueError("LightGBM doesn't support multi-label training")
+
+                dtrain = lgb.Dataset(x_train.values, label=y_train)
+                dvalid = lgb.Dataset(x_valid.values, label=y_valid)
+
+                # fit the model
+                self.model = lgb.train(
+                    self.params,
+                    dtrain,
+                    num_boost_round=num_boost_round,
+                    valid_sets=[dtrain, dvalid],
+                    valid_names=["train", "valid"],
+                    early_stopping_rounds=early_stopping_rounds,
+                    verbose_eval=verbose_eval,
+                    evals_result=evals_result,
+                    **kwargs
+                )
 
 - Override the `predict` method
     - The parameters must include the parameter `dataset`, which will be userd to get the test dataset.
     - Return the `prediction score`.
     - Please refer to `Model API <../reference/api.html#module-qlib.model.base>`_ for the parameter types of the fit method.
     - Code Example: In the following example, users need to use `LightGBM` to predict the label(such as `preds`) of test data `x_test` and return it.
-    .. code-block:: Python
 
-        def predict(self, dataset: DatasetH, **kwargs)-> pandas.Series:
-            if self.model is None:
-                raise ValueError("model is not fitted yet!")
-            x_test = dataset.prepare("test", col_set="feature", data_key=DataHandlerLP.DK_I)
-            return pd.Series(self.model.predict(x_test.values), index=x_test.index)
+        .. code-block:: Python
+
+            def predict(self, dataset: DatasetH, **kwargs)-> pandas.Series:
+                if self.model is None:
+                    raise ValueError("model is not fitted yet!")
+                x_test = dataset.prepare("test", col_set="feature", data_key=DataHandlerLP.DK_I)
+                return pd.Series(self.model.predict(x_test.values), index=x_test.index)
 
 - Override the `finetune` method (Optional)
     - This method is optional to the users. When users want to use this method on their own models, they should inherit the ``ModelFT`` base class, which includes the interface of `finetune`.
     - The parameters must include the parameter `dataset`.
     - Code Example: In the following example, users will use `LightGBM` as the model and finetune it.
-    .. code-block:: Python
 
-        def finetune(self, dataset: DatasetH, num_boost_round=10, verbose_eval=20):
-            # Based on existing model and finetune by train more rounds
-            dtrain, _ = self._prepare_data(dataset)
-            self.model = lgb.train(
-                self.params,
-                dtrain,
-                num_boost_round=num_boost_round,
-                init_model=self.model,
-                valid_sets=[dtrain],
-                valid_names=["train"],
-                verbose_eval=verbose_eval,
-            )
+        .. code-block:: Python
+
+            def finetune(self, dataset: DatasetH, num_boost_round=10, verbose_eval=20):
+                # Based on existing model and finetune by train more rounds
+                dtrain, _ = self._prepare_data(dataset)
+                self.model = lgb.train(
+                    self.params,
+                    dtrain,
+                    num_boost_round=num_boost_round,
+                    init_model=self.model,
+                    valid_sets=[dtrain],
+                    valid_names=["train"],
+                    verbose_eval=verbose_eval,
+                )
 
 Configuration File
-=======================
+==================
 
 The configuration file is described in detail in the `Workflow <../component/workflow.html#complete-example>`_ document. In order to integrate the custom model into ``Qlib``, users need to modify the "model" field in the configuration file. The configuration describes which models to use and how we can initialize it.
 
-- Example: The following example describes the `model` field of configuration file about the custom lightgbm model mentioned above, where `module_path` is the module path, `class` is the class name, and `args` is the hyperparameter passed into the __init__ method. All parameters in the field is passed to `self._params` by `\*\*kwargs` in `__init__` except `loss = mse`. 
+- Example: The following example describes the `model` field of configuration file about the custom lightgbm model mentioned above, where `module_path` is the module path, `class` is the class name, and `args` is the hyperparameter passed into the __init__ method. All parameters in the field is passed to `self._params` by `\*\*kwargs` in `__init__` except `loss = mse`.
 
-.. code-block:: YAML
-    
-    model:
-        class: LGBModel
-        module_path: qlib.contrib.model.gbdt
-        args:
-            loss: mse
-            colsample_bytree: 0.8879
-            learning_rate: 0.0421
-            subsample: 0.8789
-            lambda_l1: 205.6999
-            lambda_l2: 580.9768
-            max_depth: 8
-            num_leaves: 210
-            num_threads: 20
+    .. code-block:: YAML
+
+        model:
+            class: LGBModel
+            module_path: qlib.contrib.model.gbdt
+            args:
+                loss: mse
+                colsample_bytree: 0.8879
+                learning_rate: 0.0421
+                subsample: 0.8789
+                lambda_l1: 205.6999
+                lambda_l2: 580.9768
+                max_depth: 8
+                num_leaves: 210
+                num_threads: 20
 
 Users could find configuration file of the baselines of the ``Model`` in ``examples/benchmarks``. All the configurations of different models are listed under the corresponding model folder.
 
 Model Testing
-=====================
+=============
 Assuming that the configuration file is ``examples/benchmarks/LightGBM/workflow_config_lightgbm.yaml``, users can run the following command to test the custom model:
 
 .. code-block:: bash
@@ -136,10 +140,10 @@ Assuming that the configuration file is ``examples/benchmarks/LightGBM/workflow_
 
 .. note:: ``qrun`` is a built-in command of ``Qlib``.
 
-Also, ``Model`` can also be tested as a single module. An example has been given in ``examples/workflow_by_code.ipynb``. 
+Also, ``Model`` can also be tested as a single module. An example has been given in ``examples/workflow_by_code.ipynb``.
 
 
 Reference
-=====================
+=========
 
 To know more about ``Forecast Model``, please refer to `Forecast Model: Model Training & Prediction <../component/model.html>`_ and `Model API <../reference/api.html#module-qlib.model.base>`_.

examples/benchmarks/CatBoost/workflow_config_catboost_Alpha158_csi500.yaml

@@ -0,0 +1,72 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: CatBoostModel
+        module_path: qlib.contrib.model.catboost_model
+        kwargs:
+            loss: RMSE
+            learning_rate: 0.0421
+            subsample: 0.8789
+            max_depth: 6
+            num_leaves: 100
+            thread_count: 20
+            grow_policy: Lossguide
+            bootstrap_type: Poisson
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/CatBoost/workflow_config_catboost_Alpha360_csi500.yaml

@@ -0,0 +1,79 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors: []
+    learn_processors:
+        - class: DropnaLabel
+        - class: CSRankNorm
+          kwargs:
+              fields_group: label
+    label: ["Ref($close, -2) / Ref($close, -1) - 1"]
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: CatBoostModel
+        module_path: qlib.contrib.model.catboost_model
+        kwargs:
+            loss: RMSE
+            learning_rate: 0.0421
+            subsample: 0.8789
+            max_depth: 6
+            num_leaves: 100
+            thread_count: 20
+            grow_policy: Lossguide
+            bootstrap_type: Poisson
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha360
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha158.yaml

@@ -37,7 +37,7 @@ task:
         kwargs:
             base_model: "gbm"
             loss: mse
-            num_models: 6
+            num_models: 3
             enable_sr: True
             enable_fs: True
             alpha1: 1
@@ -53,11 +53,8 @@ task:
                 - 0.4
             sub_weights:
                 - 1
-                - 0.2
-                - 0.2
-                - 0.2
-                - 0.2
-                - 0.2
+                - 1
+                - 1
             epochs: 28
             colsample_bytree: 0.8879
             learning_rate: 0.2

examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha158_csi500.yaml

@@ -0,0 +1,97 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: DEnsembleModel
+        module_path: qlib.contrib.model.double_ensemble
+        kwargs:
+            base_model: "gbm"
+            loss: mse
+            num_models: 6
+            enable_sr: True
+            enable_fs: True
+            alpha1: 1
+            alpha2: 1
+            bins_sr: 10
+            bins_fs: 5
+            decay: 0.5
+            sample_ratios:
+                - 0.8
+                - 0.7
+                - 0.6
+                - 0.5
+                - 0.4
+            sub_weights:
+                - 1
+                - 0.2
+                - 0.2
+                - 0.2
+                - 0.2
+                - 0.2
+            epochs: 28
+            colsample_bytree: 0.8879
+            learning_rate: 0.2
+            subsample: 0.8789
+            lambda_l1: 205.6999
+            lambda_l2: 580.9768
+            max_depth: 8
+            num_leaves: 210
+            num_threads: 20
+            verbosity: -1
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha360.yaml

@@ -44,7 +44,7 @@ task:
         kwargs:
             base_model: "gbm"
             loss: mse
-            num_models: 6
+            num_models: 3
             enable_sr: True
             enable_fs: True
             alpha1: 1
@@ -60,11 +60,8 @@ task:
                 - 0.4
             sub_weights:
                 - 1
-                - 0.2
-                - 0.2
-                - 0.2
-                - 0.2
-                - 0.2
+                - 1
+                - 1
             epochs: 136
             colsample_bytree: 0.8879
             learning_rate: 0.0421

examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_Alpha360_csi500.yaml

@@ -0,0 +1,104 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors: []
+    learn_processors:
+        - class: DropnaLabel
+        - class: CSRankNorm
+          kwargs:
+              fields_group: label
+    label: ["Ref($close, -2) / Ref($close, -1) - 1"]
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: DEnsembleModel
+        module_path: qlib.contrib.model.double_ensemble
+        kwargs:
+            base_model: "gbm"
+            loss: mse
+            num_models: 6
+            enable_sr: True
+            enable_fs: True
+            alpha1: 1
+            alpha2: 1
+            bins_sr: 10
+            bins_fs: 5
+            decay: 0.5
+            sample_ratios:
+                - 0.8
+                - 0.7
+                - 0.6
+                - 0.5
+                - 0.4
+            sub_weights:
+                - 1
+                - 0.2
+                - 0.2
+                - 0.2
+                - 0.2
+                - 0.2
+            epochs: 136
+            colsample_bytree: 0.8879
+            learning_rate: 0.0421
+            subsample: 0.8789
+            lambda_l1: 205.6999
+            lambda_l2: 580.9768
+            max_depth: 8
+            num_leaves: 210
+            num_threads: 20
+            verbosity: -1
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha360
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs:
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/DoubleEnsemble/workflow_config_doubleensemble_early_stop_Alpha158.yaml

@@ -0,0 +1,95 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi300
+benchmark: &benchmark SH000300
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: DEnsembleModel
+        module_path: qlib.contrib.model.double_ensemble
+        kwargs:
+            base_model: "gbm"
+            loss: mse
+            num_models: 3
+            enable_sr: True
+            enable_fs: True
+            alpha1: 1
+            alpha2: 1
+            bins_sr: 10
+            bins_fs: 5
+            decay: 0.5
+            sample_ratios:
+                - 0.8
+                - 0.7
+                - 0.6
+                - 0.5
+                - 0.4
+            sub_weights:
+                - 1
+                - 1
+                - 1
+            epochs: 1000
+            early_stopping_rounds: 50
+            colsample_bytree: 0.8879
+            learning_rate: 0.2
+            subsample: 0.8789
+            lambda_l1: 205.6999
+            lambda_l2: 580.9768
+            max_depth: 8
+            num_leaves: 210
+            num_threads: 20
+            verbosity: -1
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/LightGBM/README.md

@@ -1,4 +1,10 @@
 # LightGBM
 * Code: [https://github.com/microsoft/LightGBM](https://github.com/microsoft/LightGBM)
 * Paper: LightGBM: A Highly Efficient Gradient Boosting
-Decision Tree. [https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf](https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf).
+Decision Tree. [https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf](https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf).
+
+
+# Introductions about the settings/configs.
+
+`workflow_config_lightgbm_multi_freq.yaml`
+- It uses data sources of different frequencies (i.e. multiple frequencies) for daily prediction.

examples/benchmarks/LightGBM/features_sample.py

@@ -5,6 +5,8 @@ from qlib.data.inst_processor import InstProcessor
 
 
 class Resample1minProcessor(InstProcessor):
+    """This processor tries to resample the data. It will reasmple the data from 1min freq to day freq by selecting a specific miniute"""
+
     def __init__(self, hour: int, minute: int, **kwargs):
         self.hour = hour
         self.minute = minute

examples/benchmarks/LightGBM/multi_freq_handler.py

@@ -29,13 +29,13 @@ class Avg15minHandler(DataHandlerLP):
         fit_end_time=None,
         process_type=DataHandlerLP.PTYPE_A,
         filter_pipe=None,
-        inst_processor=None,
+        inst_processors=None,
         **kwargs,
     ):
         infer_processors = check_transform_proc(infer_processors, fit_start_time, fit_end_time)
         learn_processors = check_transform_proc(learn_processors, fit_start_time, fit_end_time)
         data_loader = Avg15minLoader(
-            config=self.loader_config(), filter_pipe=filter_pipe, freq=freq, inst_processor=inst_processor
+            config=self.loader_config(), filter_pipe=filter_pipe, freq=freq, inst_processors=inst_processors
         )
         super().__init__(
             instruments=instruments,

examples/benchmarks/LightGBM/requirements.txt

@@ -1,3 +1,3 @@
 pandas==1.1.2
 numpy==1.21.0
-lightgbm==3.1.0
+lightgbm

examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158_csi500.yaml

@@ -0,0 +1,72 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            model: <MODEL> 
+            dataset: <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: LGBModel
+        module_path: qlib.contrib.model.gbdt
+        kwargs:
+            loss: mse
+            colsample_bytree: 0.9
+            learning_rate: 0.1
+            subsample: 0.9
+            lambda_l1: 205.6999
+            lambda_l2: 580.9768
+            max_depth: 8
+            num_leaves: 250
+            num_threads: 20
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158_multi_freq.yaml

@@ -18,7 +18,7 @@ data_handler_config: &data_handler_config
         label: day
         feature: 1min
     # with label as reference
-    inst_processor:
+    inst_processors:
         feature:
             - class: Resample1minProcessor
               module_path: features_sample.py

examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha360_csi500.yaml

@@ -0,0 +1,80 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors: []
+    learn_processors:
+        - class: DropnaLabel
+        - class: CSRankNorm
+          kwargs:
+              fields_group: label
+    label: ["Ref($close, -2) / Ref($close, -1) - 1"]
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: LGBModel
+        module_path: qlib.contrib.model.gbdt
+        kwargs:
+            loss: mse
+            colsample_bytree: 0.8879
+            learning_rate: 0.0421
+            subsample: 0.8789
+            lambda_l1: 205.6999
+            lambda_l2: 580.9768
+            max_depth: 8
+            num_leaves: 210
+            num_threads: 20
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha360
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/LightGBM/workflow_config_lightgbm_multi_freq.yaml

@@ -19,7 +19,7 @@ data_handler_config: &data_handler_config
         feature_15min: 1min
         feature_day: day
     # with label as reference
-    inst_processor:
+    inst_processors:
         feature_15min:
             - class: ResampleNProcessor
               module_path: features_resample_N.py

examples/benchmarks/Linear/workflow_config_linear_Alpha158_csi500.yaml

@@ -0,0 +1,78 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors:
+        - class: RobustZScoreNorm
+          kwargs:
+              fields_group: feature
+              clip_outlier: true
+        - class: Fillna
+          kwargs:
+              fields_group: feature
+    learn_processors:
+        - class: DropnaLabel
+        - class: CSRankNorm
+          kwargs:
+              fields_group: label
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: LinearModel
+        module_path: qlib.contrib.model.linear
+        kwargs:
+            estimator: ols
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: True
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/MLP/workflow_config_mlp_Alpha158.yaml

@@ -64,8 +64,6 @@ task:
         kwargs:
             loss: mse
             lr: 0.002
-            lr_decay: 0.96
-            lr_decay_steps: 100
             optimizer: adam
             max_steps: 8000
             batch_size: 8192

examples/benchmarks/MLP/workflow_config_mlp_Alpha158_csi500.yaml

@@ -0,0 +1,100 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors: [
+        {
+            "class" : "DropCol", 
+            "kwargs":{"col_list": ["VWAP0"]}
+        },
+        {
+             "class" : "CSZFillna", 
+             "kwargs":{"fields_group": "feature"}
+        }
+    ]
+    learn_processors: [
+        {
+            "class" : "DropCol", 
+            "kwargs":{"col_list": ["VWAP0"]}
+        },
+        {
+            "class" : "DropnaProcessor", 
+            "kwargs":{"fields_group": "feature"}
+        },
+        "DropnaLabel",
+        {
+            "class": "CSZScoreNorm", 
+            "kwargs": {"fields_group": "label"}
+        }
+    ]
+    process_type: "independent"
+
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: DNNModelPytorch
+        module_path: qlib.contrib.model.pytorch_nn
+        kwargs:
+            loss: mse
+            lr: 0.002
+            optimizer: adam
+            max_steps: 8000
+            batch_size: 8192
+            GPU: 0
+            weight_decay: 0.0002
+            pt_model_kwargs:
+              input_dim: 157
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha158
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/MLP/workflow_config_mlp_Alpha360.yaml

@@ -52,8 +52,6 @@ task:
         kwargs:
             loss: mse
             lr: 0.002
-            lr_decay: 0.96
-            lr_decay_steps: 100
             optimizer: adam
             max_steps: 8000
             batch_size: 4096

examples/benchmarks/MLP/workflow_config_mlp_Alpha360_csi500.yaml

@@ -0,0 +1,87 @@
+qlib_init:
+    provider_uri: "~/.qlib/qlib_data/cn_data"
+    region: cn
+market: &market csi500
+benchmark: &benchmark SH000905
+data_handler_config: &data_handler_config
+    start_time: 2008-01-01
+    end_time: 2020-08-01
+    fit_start_time: 2008-01-01
+    fit_end_time: 2014-12-31
+    instruments: *market
+    infer_processors:
+        - class: RobustZScoreNorm
+          kwargs:
+              fields_group: feature
+              clip_outlier: true
+        - class: Fillna
+          kwargs:
+              fields_group: feature
+    learn_processors:
+        - class: DropnaLabel
+        - class: CSRankNorm
+          kwargs:
+              fields_group: label
+    label: ["Ref($close, -2) / Ref($close, -1) - 1"]
+
+port_analysis_config: &port_analysis_config
+    strategy:
+        class: TopkDropoutStrategy
+        module_path: qlib.contrib.strategy
+        kwargs:
+            signal:
+                - <MODEL> 
+                - <DATASET>
+            topk: 50
+            n_drop: 5
+    backtest:
+        start_time: 2017-01-01
+        end_time: 2020-08-01
+        account: 100000000
+        benchmark: *benchmark
+        exchange_kwargs:
+            limit_threshold: 0.095
+            deal_price: close
+            open_cost: 0.0005
+            close_cost: 0.0015
+            min_cost: 5
+task:
+    model:
+        class: DNNModelPytorch
+        module_path: qlib.contrib.model.pytorch_nn
+        kwargs:
+            loss: mse
+            lr: 0.002
+            optimizer: adam
+            max_steps: 8000
+            batch_size: 4096
+            GPU: 0
+            pt_model_kwargs:
+              input_dim: 360
+    dataset:
+        class: DatasetH
+        module_path: qlib.data.dataset
+        kwargs:
+            handler:
+                class: Alpha360
+                module_path: qlib.contrib.data.handler
+                kwargs: *data_handler_config
+            segments:
+                train: [2008-01-01, 2014-12-31]
+                valid: [2015-01-01, 2016-12-31]
+                test: [2017-01-01, 2020-08-01]
+    record: 
+        - class: SignalRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            model: <MODEL>
+            dataset: <DATASET>
+        - class: SigAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            ana_long_short: False
+            ann_scaler: 252
+        - class: PortAnaRecord
+          module_path: qlib.workflow.record_temp
+          kwargs: 
+            config: *port_analysis_config

examples/benchmarks/README.md

@@ -20,7 +20,9 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
 > NOTE:
 > We have very limited resources to implement and finetune the models. We tried our best effort to fairly compare these models.  But some models may have greater potential than what it looks like in the table below.  Your contribution is highly welcomed to explore their potential.
 
-## Alpha158 dataset
+## Results on CSI300
+
+### Alpha158 dataset
 
 | Model Name                               | Dataset                             | IC          | ICIR        | Rank IC     | Rank ICIR   | Annualized Return | Information Ratio | Max Drawdown |
 |------------------------------------------|-------------------------------------|-------------|-------------|-------------|-------------|-------------------|-------------------|--------------|
@@ -41,10 +43,9 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
 | TFT (Bryan Lim, et al.)                  | Alpha158(with selected 20 features) | 0.0358±0.00 | 0.2160±0.03 | 0.0116±0.01 | 0.0720±0.03 | 0.0847±0.02       | 0.8131±0.19       | -0.1824±0.03 |
 | MLP                                      | Alpha158                            | 0.0376±0.00 | 0.2846±0.02 | 0.0429±0.00 | 0.3220±0.01 | 0.0895±0.02       | 1.1408±0.23       | -0.1103±0.02 |
 | LightGBM(Guolin Ke, et al.)              | Alpha158                            | 0.0448±0.00 | 0.3660±0.00 | 0.0469±0.00 | 0.3877±0.00 | 0.0901±0.00       | 1.0164±0.00       | -0.1038±0.00 |
-| DoubleEnsemble(Chuheng Zhang, et al.)    | Alpha158                            | 0.0544±0.00 | 0.4340±0.00 | 0.0523±0.00 | 0.4284±0.01 | 0.1168±0.01       | 1.3384±0.12       | -0.1036±0.01 |
+| DoubleEnsemble(Chuheng Zhang, et al.)    | Alpha158                            | 0.0521±0.00 | 0.4223±0.01 | 0.0502±0.00 | 0.4117±0.01 | 0.1158±0.01       | 1.3432±0.11       | -0.0920±0.01 |
 
-
-## Alpha360 dataset
+### Alpha360 dataset
 
 | Model Name                                | Dataset  | IC          | ICIR        | Rank IC     | Rank ICIR   | Annualized Return | Information Ratio | Max Drawdown |
 |-------------------------------------------|----------|-------------|-------------|-------------|-------------|-------------------|-------------------|--------------|
@@ -54,7 +55,7 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
 | Localformer(Juyong Jiang, et al.)         | Alpha360 | 0.0404±0.00 | 0.2932±0.04 | 0.0542±0.00 | 0.4110±0.03 | 0.0246±0.02       | 0.3211±0.21       | -0.1095±0.02 |
 | CatBoost((Liudmila Prokhorenkova, et al.) | Alpha360 | 0.0378±0.00 | 0.2714±0.00 | 0.0467±0.00 | 0.3659±0.00 | 0.0292±0.00       | 0.3781±0.00       | -0.0862±0.00 |
 | XGBoost(Tianqi Chen, et al.)              | Alpha360 | 0.0394±0.00 | 0.2909±0.00 | 0.0448±0.00 | 0.3679±0.00 | 0.0344±0.00       | 0.4527±0.02       | -0.1004±0.00 |
-| DoubleEnsemble(Chuheng Zhang, et al.)     | Alpha360 | 0.0404±0.00 | 0.3023±0.00 | 0.0495±0.00 | 0.3898±0.00 | 0.0468±0.01       | 0.6302±0.20       | -0.0860±0.01 |
+| DoubleEnsemble(Chuheng Zhang, et al.)     | Alpha360 | 0.0390±0.00 | 0.2946±0.01 | 0.0486±0.00 | 0.3836±0.01 | 0.0462±0.01       | 0.6151±0.18       | -0.0915±0.01 |
 | LightGBM(Guolin Ke, et al.)               | Alpha360 | 0.0400±0.00 | 0.3037±0.00 | 0.0499±0.00 | 0.4042±0.00 | 0.0558±0.00       | 0.7632±0.00       | -0.0659±0.00 |
 | TCN(Shaojie Bai, et al.)                  | Alpha360 | 0.0441±0.00 | 0.3301±0.02 | 0.0519±0.00 | 0.4130±0.01 | 0.0604±0.02       | 0.8295±0.34       | -0.1018±0.03 |
 | ALSTM (Yao Qin, et al.)                   | Alpha360 | 0.0497±0.00 | 0.3829±0.04 | 0.0599±0.00 | 0.4736±0.03 | 0.0626±0.02       | 0.8651±0.31       | -0.0994±0.03 |
@@ -73,12 +74,54 @@ The numbers shown below demonstrate the performance of the entire `workflow` of
 - The base model of DoubleEnsemble is LGBM.
 - The base model of TCTS is GRU.
 - About the datasets
-  - Alpha158 is a tabular dataset. There are less spatial relationships between different features. Each feature are carefully desgined by human (a.k.a feature engineering)
+  - Alpha158 is a tabular dataset. There are less spatial relationships between different features. Each feature are carefully designed by human (a.k.a feature engineering)
   - Alpha360 contains raw price and volue data without much feature engineering. There are strong strong spatial relationships between the features in the time dimension.
 - The metrics can be categorized into two
    - Signal-based evaluation:  IC, ICIR, Rank IC, Rank ICIR
+      - ![equation](https://latex.codecogs.com/gif.latex?%5Ctext%7Bcorr%7D%28%5Ctextbf%7Bx%7D%2C%5Ctextbf%7By%7D%29%3D%5Cfrac%7B%5Csum_i%20%28x_i-%5Cbar%7Bx%7D%29%28y_i-%5Cbar%7By%7D%29%7D%7B%5Csqrt%7B%5Csum_i%28x_i-%5Cbar%7Bx%7D%29%5E2%5Csum_i%28y_i-%5Cbar%7By%7D%29%5E2%7D%7D)
+      - ![equation](https://latex.codecogs.com/gif.latex?%5Ctext%7BIC%7D%5E%7B%28t%29%7D%20%3D%20%5Ctext%7Bcorr%7D%28%5Chat%7B%5Ctextbf%7By%7D%7D%5E%7B%28t%29%7D%2C%20%5Ctextbf%7Bret%7D%5E%7B%28t%29%7D%29)
+      - ![equation](https://latex.codecogs.com/gif.latex?%5Ctext%7BICIR%7D%20%3D%20%5Cfrac%20%7B%5Ctext%7Bmean%7D%28%5Ctextbf%7BIC%7D%29%7D%20%7B%5Ctext%7Bstd%7D%28%5Ctextbf%7BIC%7D%29%7D)
+      - ![equation](https://latex.codecogs.com/gif.latex?%5Ctext%7BRank%20IC%7D%5E%7B%28t%29%7D%20%3D%20%5Ctext%7Bcorr%7D%28%5Ctext%7Brank%7D%28%5Chat%7B%5Ctextbf%7By%7D%7D%5E%7B%28t%29%7D%29%2C%20%5Ctext%7Brank%7D%28%5Ctextbf%7Bret%7D%5E%7B%28t%29%7D%29%29)
+      - ![equation](https://latex.codecogs.com/gif.latex?%5Ctext%7BRank%20ICIR%7D%20%3D%20%5Cfrac%20%7B%5Ctext%7Bmean%7D%28%5Ctextbf%7BRank%20IC%7D%29%7D%20%7B%5Ctext%7Bstd%7D%28%5Ctextbf%7BRankIC%7D%29%7D)
    - Portfolio-based metrics:  Annualized Return, Information Ratio, Max Drawdown
 
+## Results on CSI500
+The results on CSI500 is not complete. PR's for models on csi500 are welcome!
+
+Transfer previous models in CSI300 to CSI500 is quite easy.  You can try models with just a few commands below.
+```
+cd examples/benchmarks/LightGBM
+pip install -r requirements.txt
+
+# create new config and set the benchmark to csi500
+cp workflow_config_lightgbm_Alpha158.yaml workflow_config_lightgbm_Alpha158_csi500.yaml
+sed -i "s/csi300/csi500/g"  workflow_config_lightgbm_Alpha158_csi500.yaml
+sed -i "s/SH000300/SH000905/g"  workflow_config_lightgbm_Alpha158_csi500.yaml
+
+# you can either run the model once
+qrun workflow_config_lightgbm_Alpha158_csi500.yaml
+
+# or run it for multiple times automatically and get the summarized results.
+cd  ../../
+python run_all_model.py run 3 lightgbm Alpha158 csi500  # for models with randomness.  please run it for 20 times.
+```
+
+### Alpha158 dataset
+| Model Name | Dataset  | IC          | ICIR        | Rank IC     | Rank ICIR   | Annualized Return | Information Ratio | Max Drawdown |
+|------------|----------|-------------|-------------|-------------|-------------|-------------------|-------------------|--------------|
+| Linear     | Alpha158 | 0.0332±0.00 | 0.3044±0.00 | 0.0462±0.00 | 0.4326±0.00 | 0.0382±0.00       | 0.1723±0.00       | -0.4876±0.00 |
+| MLP        | Alpha158 | 0.0229±0.01 | 0.2181±0.05 | 0.0360±0.00 | 0.3409±0.02 | 0.0043±0.02       | 0.0602±0.27       | -0.2184±0.04 |
+| LightGBM   | Alpha158 | 0.0399±0.00 | 0.4065±0.00 | 0.0482±0.00 | 0.5101±0.00 | 0.1284±0.00       | 1.5650±0.00       | -0.0635±0.00 |
+| CatBoost   | Alpha158 | 0.0345±0.00 | 0.2855±0.00 | 0.0417±0.00 | 0.3740±0.00 | 0.0496±0.00       | 0.5977±0.00       | -0.1496±0.00 |
+| DoubleEnsemble  | Alpha158 | 0.0380±0.00 | 0.3659±0.00 | 0.0442±0.00 | 0.4324±0.00 | 0.0382±0.00       | 0.1723±0.00       | -0.4876±0.00 |
+
+### Alpha360 dataset
+| Model Name | Dataset  | IC          | ICIR        | Rank IC     | Rank ICIR   | Annualized Return | Information Ratio | Max Drawdown |
+|------------|----------|-------------|-------------|-------------|-------------|-------------------|-------------------|--------------|
+| MLP        | Alpha360 | 0.0258±0.00 | 0.2021±0.02 | 0.0426±0.00 | 0.3840±0.02 | 0.0022±0.02       | 0.0301±0.26       | -0.2064±0.02 |
+| LightGBM   | Alpha360 | 0.0400±0.00 | 0.3605±0.00 | 0.0536±0.00 | 0.5431±0.00 | 0.0505±0.00       | 0.7658±0.02       | -0.1880±0.00 |
+| CatBoost   | Alpha360 | 0.0382±0.00 | 0.3229±0.00 | 0.0489±0.00 | 0.4649±0.00 | 0.0297±0.00       | 0.4227±0.02       | -0.1499±0.01 |
+| DoubleEnsemble  | Alpha360 | 0.0361±0.00 | 0.3092±0.00 | 0.0499±0.00 | 0.4793±0.00 | 0.0382±0.00       | 0.1723±0.02       | -0.4876±0.00 |
 
 # Contributing
 
@@ -95,3 +138,10 @@ If you want to contribute your new models, you can follow the steps below.
 5. Update the info in the index page in the [news list](https://github.com/microsoft/qlib#newspaper-whats-new----sparkling_heart) and [model list](https://github.com/microsoft/qlib#quant-model-paper-zoo).
 
 Finally, you can send PR for review. ([here is an example](https://github.com/microsoft/qlib/pull/1040))
+
+
+# FAQ
+
+Q: What's the difference between models with name `*.py` and `*_ts.py`?
+
+A: Models with name `*_ts.py` are designed for `TSDatasetH` (`TSDatasetH` will create time-series automatically from tabular data).  Models with name `*.py` are designed for `DatasetH` (`DatasetH` is usually used in tabular data.  But users still can apply time-series models on tabular datasets if the columns has time-series relationships). 

examples/benchmarks/TRA/Reports.ipynb

@@ -25,59 +25,65 @@
     "import seaborn as sns\n",
     "import matplotlib.pyplot as plt\n",
     "import matplotlib\n",
-    "sns.set(style='white')\n",
-    "matplotlib.rcParams['pdf.fonttype'] = 42\n",
-    "matplotlib.rcParams['ps.fonttype'] = 42\n",
+    "\n",
+    "sns.set(style=\"white\")\n",
+    "matplotlib.rcParams[\"pdf.fonttype\"] = 42\n",
+    "matplotlib.rcParams[\"ps.fonttype\"] = 42\n",
     "\n",
     "from tqdm.auto import tqdm\n",
     "from joblib import Parallel, delayed\n",
     "\n",
+    "\n",
     "def func(x, N=80):\n",
     "    ret = x.ret.copy()\n",
     "    x = x.rank(pct=True)\n",
-    "    x['ret'] = ret\n",
+    "    x[\"ret\"] = ret\n",
     "    diff = x.score.sub(x.label)\n",
-    "    r = x.nlargest(N, columns='score').ret.mean()\n",
-    "    r -= x.nsmallest(N, columns='score').ret.mean()\n",
-    "    return pd.Series({\n",
-    "        'MSE': diff.pow(2).mean(), \n",
-    "        'MAE': diff.abs().mean(), \n",
-    "        'IC': x.score.corr(x.label),\n",
-    "        'R': r\n",
-    "    })\n",
-    "    \n",
+    "    r = x.nlargest(N, columns=\"score\").ret.mean()\n",
+    "    r -= x.nsmallest(N, columns=\"score\").ret.mean()\n",
+    "    return pd.Series(\n",
+    "        {\n",
+    "            \"MSE\": diff.pow(2).mean(),\n",
+    "            \"MAE\": diff.abs().mean(),\n",
+    "            \"IC\": x.score.corr(x.label),\n",
+    "            \"R\": r,\n",
+    "        }\n",
+    "    )\n",
+    "\n",
+    "\n",
     "ret = pd.read_pickle(\"data/ret.pkl\").clip(-0.1, 0.1)\n",
+    "\n",
+    "\n",
     "def backtest(fname, **kwargs):\n",
-    "    pred = pd.read_pickle(fname).loc['2018-09-21':'2020-06-30']  # test period\n",
-    "    pred['ret'] = ret\n",
+    "    pred = pd.read_pickle(fname).loc[\"2018-09-21\":\"2020-06-30\"]  # test period\n",
+    "    pred[\"ret\"] = ret\n",
     "    dates = pred.index.unique(level=0)\n",
     "    res = Parallel(n_jobs=-1)(delayed(func)(pred.loc[d], **kwargs) for d in dates)\n",
-    "    res = {\n",
-    "       dates[i]: res[i]\n",
-    "       for i in range(len(dates))\n",
-    "    }\n",
+    "    res = {dates[i]: res[i] for i in range(len(dates))}\n",
     "    res = pd.DataFrame(res).T\n",
-    "    r = res['R'].copy()\n",
+    "    r = res[\"R\"].copy()\n",
     "    r.index = pd.to_datetime(r.index)\n",
     "    r = r.reindex(pd.date_range(r.index[0], r.index[-1])).fillna(0)  # paper use 365 days\n",
     "    return {\n",
-    "        'MSE': res['MSE'].mean(),\n",
-    "        'MAE': res['MAE'].mean(),\n",
-    "        'IC': res['IC'].mean(),\n",
-    "        'ICIR': res['IC'].mean()/res['IC'].std(),\n",
-    "        'AR': r.mean()*365,\n",
-    "        'AV': r.std()*365**0.5,\n",
-    "        'SR': r.mean()/r.std()*365**0.5,\n",
-    "        'MDD': (r.cumsum().cummax() - r.cumsum()).max()\n",
+    "        \"MSE\": res[\"MSE\"].mean(),\n",
+    "        \"MAE\": res[\"MAE\"].mean(),\n",
+    "        \"IC\": res[\"IC\"].mean(),\n",
+    "        \"ICIR\": res[\"IC\"].mean() / res[\"IC\"].std(),\n",
+    "        \"AR\": r.mean() * 365,\n",
+    "        \"AV\": r.std() * 365**0.5,\n",
+    "        \"SR\": r.mean() / r.std() * 365**0.5,\n",
+    "        \"MDD\": (r.cumsum().cummax() - r.cumsum()).max(),\n",
     "    }, r\n",
     "\n",
+    "\n",
     "def fmt(x, p=3, scale=1, std=False):\n",
-    "    _fmt = '{:.%df}'%p\n",
+    "    _fmt = \"{:.%df}\" % p\n",
     "    string = _fmt.format((x.mean() if not isinstance(x, (float, np.floating)) else x) * scale)\n",
     "    if std and len(x) > 1:\n",
-    "        string += ' ('+_fmt.format(x.std()*scale)+')'\n",
+    "        string += \" (\" + _fmt.format(x.std() * scale) + \")\"\n",
     "    return string\n",
     "\n",
+    "\n",
     "def backtest_multi(files, **kwargs):\n",
     "    res = []\n",
     "    pnl = []\n",
@@ -88,14 +94,14 @@
     "    res = pd.DataFrame(res)\n",
     "    pnl = pd.concat(pnl, axis=1)\n",
     "    return {\n",
-    "        'MSE': fmt(res['MSE'], std=True),\n",
-    "        'MAE': fmt(res['MAE'], std=True),\n",
-    "        'IC': fmt(res['IC']),\n",
-    "        'ICIR': fmt(res['ICIR']),\n",
-    "        'AR': fmt(res['AR'], scale=100, p=1)+'%',\n",
-    "        'VR': fmt(res['AV'], scale=100, p=1)+'%',\n",
-    "        'SR': fmt(res['SR']),\n",
-    "        'MDD': fmt(res['MDD'], scale=100, p=1)+'%'\n",
+    "        \"MSE\": fmt(res[\"MSE\"], std=True),\n",
+    "        \"MAE\": fmt(res[\"MAE\"], std=True),\n",
+    "        \"IC\": fmt(res[\"IC\"]),\n",
+    "        \"ICIR\": fmt(res[\"ICIR\"]),\n",
+    "        \"AR\": fmt(res[\"AR\"], scale=100, p=1) + \"%\",\n",
+    "        \"VR\": fmt(res[\"AV\"], scale=100, p=1) + \"%\",\n",
+    "        \"SR\": fmt(res[\"SR\"]),\n",
+    "        \"MDD\": fmt(res[\"MDD\"], scale=100, p=1) + \"%\",\n",
     "    }, pnl"
    ]
   },
@@ -124,16 +130,20 @@
    "outputs": [],
    "source": [
     "exps = {\n",
-    "    'Linear': ['output/Linear/pred.pkl'],\n",
-    "    'LightGBM': ['output/GBDT/lr0.05_leaves128/pred.pkl'],\n",
-    "    'MLP': glob.glob('output/search/MLP/hs128_bs512_do0.3_lr0.001_seed*/pred.pkl'),\n",
-    "    'SFM': glob.glob('output/search/SFM/hs32_bs512_do0.5_lr0.001_seed*/pred.pkl'),\n",
-    "    'ALSTM': glob.glob('output/search/LSTM_Attn/hs256_bs1024_do0.1_lr0.0002_seed*/pred.pkl'),\n",
-    "    'Trans.': glob.glob('output/search/Transformer/head4_hs64_bs1024_do0.1_lr0.0002_seed*/pred.pkl'),\n",
-    "    'ALSTM+TS':glob.glob('output/LSTM_Attn_TS/hs256_bs1024_do0.1_lr0.0002_seed*/pred.pkl'),\n",
-    "    'Trans.+TS':glob.glob('output/Transformer_TS/head4_hs64_bs1024_do0.1_lr0.0002_seed*/pred.pkl'),\n",
-    "    'ALSTM+TRA(Ours)': glob.glob('output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl'),\n",
-    "    'Trans.+TRA(Ours)': glob.glob('output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb1.0_head4_hs64_bs512_do0.1_lr0.0005_seed*/pred.pkl')\n",
+    "    \"Linear\": [\"output/Linear/pred.pkl\"],\n",
+    "    \"LightGBM\": [\"output/GBDT/lr0.05_leaves128/pred.pkl\"],\n",
+    "    \"MLP\": glob.glob(\"output/search/MLP/hs128_bs512_do0.3_lr0.001_seed*/pred.pkl\"),\n",
+    "    \"SFM\": glob.glob(\"output/search/SFM/hs32_bs512_do0.5_lr0.001_seed*/pred.pkl\"),\n",
+    "    \"ALSTM\": glob.glob(\"output/search/LSTM_Attn/hs256_bs1024_do0.1_lr0.0002_seed*/pred.pkl\"),\n",
+    "    \"Trans.\": glob.glob(\"output/search/Transformer/head4_hs64_bs1024_do0.1_lr0.0002_seed*/pred.pkl\"),\n",
+    "    \"ALSTM+TS\": glob.glob(\"output/LSTM_Attn_TS/hs256_bs1024_do0.1_lr0.0002_seed*/pred.pkl\"),\n",
+    "    \"Trans.+TS\": glob.glob(\"output/Transformer_TS/head4_hs64_bs1024_do0.1_lr0.0002_seed*/pred.pkl\"),\n",
+    "    \"ALSTM+TRA(Ours)\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl\"\n",
+    "    ),\n",
+    "    \"Trans.+TRA(Ours)\": glob.glob(\n",
+    "        \"output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb1.0_head4_hs64_bs512_do0.1_lr0.0005_seed*/pred.pkl\"\n",
+    "    ),\n",
     "}"
    ]
   },
@@ -160,14 +170,8 @@
     }
    ],
    "source": [
-    "res = {\n",
-    "    name: backtest_multi(exps[name])\n",
-    "    for name in tqdm(exps)\n",
-    "}\n",
-    "report = pd.DataFrame({\n",
-    "    k: v[0]\n",
-    "    for k, v in res.items()\n",
-    "}).T"
+    "res = {name: backtest_multi(exps[name]) for name in tqdm(exps)}\n",
+    "report = pd.DataFrame({k: v[0] for k, v in res.items()}).T"
    ]
   },
   {
@@ -385,24 +389,40 @@
     }
    ],
    "source": [
-    "df = pd.read_pickle('output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb0.0_head4_hs64_bs512_do0.1_lr0.0005_seed1000/pred.pkl')\n",
-    "code = 'SH600157'\n",
-    "date = '2018-09-28'\n",
+    "df = pd.read_pickle(\n",
+    "    \"output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb0.0_head4_hs64_bs512_do0.1_lr0.0005_seed1000/pred.pkl\"\n",
+    ")\n",
+    "code = \"SH600157\"\n",
+    "date = \"2018-09-28\"\n",
     "lookbackperiod = 50\n",
     "\n",
     "prob = df.iloc[:, -3:].loc(axis=0)[:, code].reset_index(level=1, drop=True).loc[date:].iloc[:lookbackperiod]\n",
-    "pred = df.loc[:,[\"score_0\",\"score_1\",\"score_2\",\"label\"]].loc(axis=0)[:, code].reset_index(level=1, drop=True).loc[date:].iloc[:lookbackperiod]\n",
-    "e_all = pred.iloc[:,:-1].sub(pred.iloc[:,-1], axis=0).pow(2)\n",
+    "pred = (\n",
+    "    df.loc[:, [\"score_0\", \"score_1\", \"score_2\", \"label\"]]\n",
+    "    .loc(axis=0)[:, code]\n",
+    "    .reset_index(level=1, drop=True)\n",
+    "    .loc[date:]\n",
+    "    .iloc[:lookbackperiod]\n",
+    ")\n",
+    "e_all = pred.iloc[:, :-1].sub(pred.iloc[:, -1], axis=0).pow(2)\n",
     "e_all = e_all.sub(e_all.min(axis=1), axis=0)\n",
-    "e_all.columns = [r'$\\theta_%d$'%d for d in range(1, 4)]\n",
+    "e_all.columns = [r\"$\\theta_%d$\" % d for d in range(1, 4)]\n",
     "prob = pd.Series(np.argmax(prob.values, axis=1), index=prob.index).rolling(7).mean().round()\n",
     "\n",
     "fig, axes = plt.subplots(1, 2, figsize=(7, 3))\n",
-    "e_all.plot(ax=axes[0], xlabel='', rot=30)\n",
-    "prob.plot(ax=axes[1], xlabel='', rot=30, color='red', linestyle='None', marker='^', markersize=5)\n",
+    "e_all.plot(ax=axes[0], xlabel=\"\", rot=30)\n",
+    "prob.plot(\n",
+    "    ax=axes[1],\n",
+    "    xlabel=\"\",\n",
+    "    rot=30,\n",
+    "    color=\"red\",\n",
+    "    linestyle=\"None\",\n",
+    "    marker=\"^\",\n",
+    "    markersize=5,\n",
+    ")\n",
     "plt.yticks(np.array([0, 1, 2]), e_all.columns.values)\n",
-    "axes[0].set_ylabel('Predictor Loss')\n",
-    "axes[1].set_ylabel('Router Selection')\n",
+    "axes[0].set_ylabel(\"Predictor Loss\")\n",
+    "axes[1].set_ylabel(\"Router Selection\")\n",
     "plt.tight_layout()\n",
     "# plt.savefig('select.pdf', bbox_inches='tight')\n",
     "plt.show()"
@@ -428,10 +448,18 @@
    "outputs": [],
    "source": [
     "exps = {\n",
-    "    'Random': glob.glob('output/search/LSTM_Attn_tra/K10_traHs16_traSrcNONE_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl'),\n",
-    "    'LR': glob.glob('output/search/LSTM_Attn_tra/K10_traHs16_traSrcLR_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl'),\n",
-    "    'TPE': glob.glob('output/search/LSTM_Attn_tra/K10_traHs16_traSrcTPE_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl'),\n",
-    "    'LR+TPE': glob.glob('output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl')\n",
+    "    \"Random\": glob.glob(\n",
+    "        \"output/search/LSTM_Attn_tra/K10_traHs16_traSrcNONE_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl\"\n",
+    "    ),\n",
+    "    \"LR\": glob.glob(\n",
+    "        \"output/search/LSTM_Attn_tra/K10_traHs16_traSrcLR_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl\"\n",
+    "    ),\n",
+    "    \"TPE\": glob.glob(\n",
+    "        \"output/search/LSTM_Attn_tra/K10_traHs16_traSrcTPE_traLamb1.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl\"\n",
+    "    ),\n",
+    "    \"LR+TPE\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/pred.pkl\"\n",
+    "    ),\n",
     "}"
    ]
   },
@@ -456,14 +484,8 @@
     }
    ],
    "source": [
-    "res = {\n",
-    "    name: backtest_multi(exps[name])\n",
-    "    for name in tqdm(exps)\n",
-    "}\n",
-    "report = pd.DataFrame({\n",
-    "    k: v[0]\n",
-    "    for k, v in res.items()\n",
-    "}).T"
+    "res = {name: backtest_multi(exps[name]) for name in tqdm(exps)}\n",
+    "report = pd.DataFrame({k: v[0] for k, v in res.items()}).T"
    ]
   },
   {
@@ -597,18 +619,22 @@
     }
    ],
    "source": [
-    "a = pd.read_pickle('output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb0.0_head4_hs64_bs512_do0.1_lr0.0005_seed3000/pred.pkl')\n",
-    "b = pd.read_pickle('output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb2.0_head4_hs64_bs512_do0.1_lr0.0005_seed3000/pred.pkl')\n",
+    "a = pd.read_pickle(\n",
+    "    \"output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb0.0_head4_hs64_bs512_do0.1_lr0.0005_seed3000/pred.pkl\"\n",
+    ")\n",
+    "b = pd.read_pickle(\n",
+    "    \"output/search/finetune/Transformer_tra/K3_traHs16_traSrcLR_TPE_traLamb2.0_head4_hs64_bs512_do0.1_lr0.0005_seed3000/pred.pkl\"\n",
+    ")\n",
     "a = a.iloc[:, -3:]\n",
     "b = b.iloc[:, -3:]\n",
     "b = np.eye(3)[b.values.argmax(axis=1)]\n",
     "a = np.eye(3)[a.values.argmax(axis=1)]\n",
     "\n",
-    "res = pd.DataFrame({\n",
-    "    'with OT': b.sum(axis=0) / b.sum(),\n",
-    "    'without OT': a.sum(axis=0)/ a.sum()  \n",
-    "},index=[r'$\\theta_1$',r'$\\theta_2$',r'$\\theta_3$'])\n",
-    "res.plot.bar(rot=30, figsize=(5, 4), color=['b', 'g'])\n",
+    "res = pd.DataFrame(\n",
+    "    {\"with OT\": b.sum(axis=0) / b.sum(), \"without OT\": a.sum(axis=0) / a.sum()},\n",
+    "    index=[r\"$\\theta_1$\", r\"$\\theta_2$\", r\"$\\theta_3$\"],\n",
+    ")\n",
+    "res.plot.bar(rot=30, figsize=(5, 4), color=[\"b\", \"g\"])\n",
     "del a, b"
    ]
   },
@@ -633,11 +659,19 @@
    "outputs": [],
    "source": [
     "exps = {\n",
-    "    'K=1': glob.glob('output/search/LSTM_Attn/hs256_bs1024_do0.1_lr0.0002_seed*/info.json'),\n",
-    "    'K=3': glob.glob('output/search/finetune/LSTM_Attn_tra/K3_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json'),\n",
-    "    'K=5': glob.glob('output/search/finetune/LSTM_Attn_tra/K5_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json'),\n",
-    "    'K=10': glob.glob('output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json'),\n",
-    "    'K=20': glob.glob('output/search/finetune/LSTM_Attn_tra/K20_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json')\n",
+    "    \"K=1\": glob.glob(\"output/search/LSTM_Attn/hs256_bs1024_do0.1_lr0.0002_seed*/info.json\"),\n",
+    "    \"K=3\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K3_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json\"\n",
+    "    ),\n",
+    "    \"K=5\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K5_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json\"\n",
+    "    ),\n",
+    "    \"K=10\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K10_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json\"\n",
+    "    ),\n",
+    "    \"K=20\": glob.glob(\n",
+    "        \"output/search/finetune/LSTM_Attn_tra/K20_traHs16_traSrcLR_TPE_traLamb2.0_hs256_bs1024_do0.1_lr0.0001_seed*/info.json\"\n",
+    "    ),\n",
     "}"
    ]
   },
@@ -649,16 +683,11 @@
    "source": [
     "report = dict()\n",
     "for k, v in exps.items():\n",
-    "    \n",
     "    tmp = dict()\n",
     "    for fname in v:\n",
     "        with open(fname) as f:\n",
     "            info = json.load(f)\n",
-    "        tmp[fname] = (\n",
-    "        {\n",
-    "            \"IC\":info[\"metric\"][\"IC\"],\n",
-    "            \"MSE\":info[\"metric\"][\"MSE\"]\n",
-    "        })\n",
+    "        tmp[fname] = {\"IC\": info[\"metric\"][\"IC\"], \"MSE\": info[\"metric\"][\"MSE\"]}\n",
     "    tmp = pd.DataFrame(tmp).T\n",
     "    report[k] = tmp.mean()\n",
     "report = pd.DataFrame(report).T"
@@ -681,13 +710,14 @@
     }
    ],
    "source": [
-    "fig, axes = plt.subplots(1, 2, figsize=(6,3)); axes = axes.flatten()\n",
-    "report['IC'].plot.bar(rot=30, ax=axes[0])\n",
+    "fig, axes = plt.subplots(1, 2, figsize=(6, 3))\n",
+    "axes = axes.flatten()\n",
+    "report[\"IC\"].plot.bar(rot=30, ax=axes[0])\n",
     "axes[0].set_ylim(0.045, 0.062)\n",
-    "axes[0].set_title('IC performance')\n",
-    "report['MSE'].astype(float).plot.bar(rot=30, ax=axes[1], color='green')\n",
+    "axes[0].set_title(\"IC performance\")\n",
+    "report[\"MSE\"].astype(float).plot.bar(rot=30, ax=axes[1], color=\"green\")\n",
     "axes[1].set_ylim(0.155, 0.1585)\n",
-    "axes[1].set_title('MSE performance')\n",
+    "axes[1].set_title(\"MSE performance\")\n",
     "plt.tight_layout()\n",
     "# plt.savefig('sensitivity.pdf')"
    ]

examples/benchmarks_dynamic/DDG-DA/README.md

@@ -28,6 +28,8 @@ The default forecasting models are `Linear`. Users can choose other forecasting
 The results of related methods in Qlib's public dataset can be found [here](../)
 
 # Requirements
-Here is the minimal hardware requirements to run the ``workflow.py`` of DDG-DA.
+Here are the minimal hardware requirements to run the ``workflow.py`` of DDG-DA.
 * Memory: 45G
 * Disk: 4G
+
+Pytorch with CPU & RAM will be enough for this example.

examples/benchmarks_dynamic/DDG-DA/vis_data.py

@@ -0,0 +1,107 @@
+import pickle
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+sns.set(color_codes=True)
+plt.rcParams["font.sans-serif"] = "SimHei"
+plt.rcParams["axes.unicode_minus"] = False
+from tqdm.auto import tqdm
+
+# tqdm.pandas()  # for progress_apply
+# %matplotlib inline
+# %load_ext autoreload
+
+
+# # Meta Input
+
+# +
+with open("./internal_data_s20.pkl", "rb") as f:
+    data = pickle.load(f)
+
+data.data_ic_df.columns.names = ["start_date", "end_date"]
+
+data_sim = data.data_ic_df.droplevel(axis=1, level="end_date")
+
+data_sim.index.name = "test datetime"
+# -
+
+plt.figure(figsize=(40, 20))
+sns.heatmap(data_sim)
+
+plt.figure(figsize=(40, 20))
+sns.heatmap(data_sim.rolling(20).mean())
+
+# # Meta Model
+
+from qlib import auto_init
+
+auto_init()
+from qlib.workflow import R
+
+exp = R.get_exp(experiment_name="DDG-DA")
+meta_rec = exp.list_recorders(rtype="list", max_results=1)[0]
+meta_m = meta_rec.load_object("model")
+
+pd.DataFrame(meta_m.tn.twm.linear.weight.detach().numpy()).T[0].plot()
+
+pd.DataFrame(meta_m.tn.twm.linear.weight.detach().numpy()).T[0].rolling(5).mean().plot()
+
+# # Meta Output
+
+# +
+with open("./tasks_s20.pkl", "rb") as f:
+    tasks = pickle.load(f)
+
+task_df = {}
+for t in tasks:
+    test_seg = t["dataset"]["kwargs"]["segments"]["test"]
+    if None not in test_seg:
+        # The last rolling is skipped.
+        task_df[test_seg] = t["reweighter"].time_weight
+task_df = pd.concat(task_df)
+
+task_df.index.names = ["OS_start", "OS_end", "IS_start", "IS_end"]
+task_df = task_df.droplevel(["OS_end", "IS_end"])
+task_df = task_df.unstack("OS_start")
+# -
+
+plt.figure(figsize=(40, 20))
+sns.heatmap(task_df.T)
+
+plt.figure(figsize=(40, 20))
+sns.heatmap(task_df.rolling(10).mean().T)
+
+# # Sub Models
+#
+# NOTE:
+# - this section assumes that the model is Linear model!!
+# - Other models does not support this analysis
+
+exp = R.get_exp(experiment_name="rolling_ds")
+
+
+def show_linear_weight(exp):
+    coef_df = {}
+    for r in exp.list_recorders("list"):
+        t = r.load_object("task")
+        if None in t["dataset"]["kwargs"]["segments"]["test"]:
+            continue
+        m = r.load_object("params.pkl")
+        coef_df[t["dataset"]["kwargs"]["segments"]["test"]] = pd.Series(m.coef_)
+
+    coef_df = pd.concat(coef_df)
+
+    coef_df.index.names = ["test_start", "test_end", "coef_idx"]
+
+    coef_df = coef_df.droplevel("test_end").unstack("coef_idx").T
+
+    plt.figure(figsize=(40, 20))
+    sns.heatmap(coef_df)
+    plt.show()
+
+
+show_linear_weight(R.get_exp(experiment_name="rolling_ds"))
+
+show_linear_weight(R.get_exp(experiment_name="rolling_models"))

examples/benchmarks_dynamic/DDG-DA/workflow.py

@@ -10,8 +10,10 @@ import pandas as pd
 import fire
 import sys
 import pickle
+from typing import Optional
 from qlib import auto_init
 from qlib.model.trainer import TrainerR
+from qlib.typehint import Literal
 from qlib.utils import init_instance_by_config
 from qlib.workflow import R
 from qlib.tests.data import GetData
@@ -30,7 +32,33 @@ class DDGDA:
     - `rm -r mlruns`
     """
 
-    def __init__(self, sim_task_model="linear", forecast_model="linear"):
+    def __init__(
+        self,
+        sim_task_model: Literal["linear", "gbdt"] = "linear",
+        forecast_model: Literal["linear", "gbdt"] = "linear",
+        h_path: Optional[str] = None,
+        test_end: Optional[str] = None,
+        train_start: Optional[str] = None,
+        meta_1st_train_end: Optional[str] = None,
+        task_ext_conf: Optional[dict] = None,
+        alpha: float = 0.0,
+        proxy_hd: str = "handler_proxy.pkl",
+    ):
+        """
+
+        Parameters
+        ----------
+
+        train_start: Optional[str]
+            the start datetime for data.  It is used in training start time (for both tasks & meta learing)
+        test_end: Optional[str]
+            the end datetime for data. It is used in test end time
+        meta_1st_train_end: Optional[str]
+            the datetime of training end of the first meta_task
+        alpha: float
+            Setting the L2 regularization for ridge
+            The `alpha` is only passed to MetaModelDS (it is not passed to sim_task_model currently..)
+        """
         self.step = 20
         # NOTE:
         # the horizon must match the meaning in the base task template
@@ -38,10 +66,19 @@ class DDGDA:
         self.meta_exp_name = "DDG-DA"
         self.sim_task_model = sim_task_model  # The model to capture the distribution of data.
         self.forecast_model = forecast_model  # downstream forecasting models' type
+        self.rb_kwargs = {
+            "h_path": h_path,
+            "test_end": test_end,
+            "train_start": train_start,
+            "task_ext_conf": task_ext_conf,
+        }
+        self.alpha = alpha
+        self.meta_1st_train_end = meta_1st_train_end
+        self.proxy_hd = proxy_hd
 
     def get_feature_importance(self):
         # this must be lightGBM, because it needs to get the feature importance
-        rb = RollingBenchmark(model_type="gbdt")
+        rb = RollingBenchmark(model_type="gbdt", **self.rb_kwargs)
         task = rb.basic_task()
 
         with R.start(experiment_name="feature_importance"):
@@ -69,7 +106,7 @@ class DDGDA:
         fi = self.get_feature_importance()
         col_selected = fi.nlargest(topk)
 
-        rb = RollingBenchmark(model_type=self.sim_task_model)
+        rb = RollingBenchmark(model_type=self.sim_task_model, **self.rb_kwargs)
         task = rb.basic_task()
         dataset = init_instance_by_config(task["dataset"])
         prep_ds = dataset.prepare(slice(None), col_set=["feature", "label"], data_key=DataHandlerLP.DK_L)
@@ -96,7 +133,7 @@ class DDGDA:
                 "kwargs": {"config": DIRNAME / "fea_label_df.pkl"},
             }
         )
-        handler.to_pickle(DIRNAME / "handler_proxy.pkl", dump_all=True)
+        handler.to_pickle(DIRNAME / self.proxy_hd, dump_all=True)
 
     @property
     def _internal_data_path(self):
@@ -108,7 +145,7 @@ class DDGDA:
         This function will dump the input data for meta model
         """
         # According to the experiments, the choice of the model type is very important for achieving good results
-        rb = RollingBenchmark(model_type=self.sim_task_model)
+        rb = RollingBenchmark(model_type=self.sim_task_model, **self.rb_kwargs)
         sim_task = rb.basic_task()
 
         if self.sim_task_model == "gbdt":
@@ -122,24 +159,27 @@ class DDGDA:
         with self._internal_data_path.open("wb") as f:
             pickle.dump(internal_data, f)
 
-    def train_meta_model(self):
+    def train_meta_model(self, fill_method="max"):
         """
         training a meta model based on a simplified linear proxy model;
         """
 
         # 1) leverage the simplified proxy forecasting model to train meta model.
         # - Only the dataset part is important, in current version of meta model will integrate the
-        rb = RollingBenchmark(model_type=self.sim_task_model)
+        rb = RollingBenchmark(model_type=self.sim_task_model, **self.rb_kwargs)
         sim_task = rb.basic_task()
+        train_start = self.rb_kwargs.get("train_start", "2008-01-01")
+        train_end = "2010-12-31" if self.meta_1st_train_end is None else self.meta_1st_train_end
+        test_start = (pd.Timestamp(train_end) + pd.Timedelta(days=1)).strftime("%Y-%m-%d")
         proxy_forecast_model_task = {
             # "model": "qlib.contrib.model.linear.LinearModel",
             "dataset": {
                 "class": "qlib.data.dataset.DatasetH",
                 "kwargs": {
-                    "handler": f"file://{(DIRNAME / 'handler_proxy.pkl').absolute()}",
+                    "handler": f"file://{(DIRNAME / self.proxy_hd).absolute()}",
                     "segments": {
-                        "train": ("2008-01-01", "2010-12-31"),
-                        "test": ("2011-01-01", sim_task["dataset"]["kwargs"]["segments"]["test"][1]),
+                        "train": (train_start, train_end),
+                        "test": (test_start, sim_task["dataset"]["kwargs"]["segments"]["test"][1]),
                     },
                 },
             },
@@ -156,7 +196,7 @@ class DDGDA:
             segments=0.62,  # keep test period consistent with the dataset yaml
             trunc_days=1 + self.horizon,
             hist_step_n=30,
-            fill_method="max",
+            fill_method=fill_method,
             rolling_ext_days=0,
         )
         # NOTE:
@@ -165,12 +205,15 @@ class DDGDA:
         # So the misalignment will not affect the effectiveness of the method.
         with self._internal_data_path.open("rb") as f:
             internal_data = pickle.load(f)
+
         md = MetaDatasetDS(exp_name=internal_data, **kwargs)
 
         # 3) train and logging meta model
         with R.start(experiment_name=self.meta_exp_name):
             R.log_params(**kwargs)
-            mm = MetaModelDS(step=self.step, hist_step_n=kwargs["hist_step_n"], lr=0.001, max_epoch=200, seed=43)
+            mm = MetaModelDS(
+                step=self.step, hist_step_n=kwargs["hist_step_n"], lr=0.001, max_epoch=100, seed=43, alpha=self.alpha
+            )
             mm.fit(md)
             R.save_objects(model=mm)
 
@@ -203,7 +246,7 @@ class DDGDA:
         hist_step_n = int(param["hist_step_n"])
         fill_method = param.get("fill_method", "max")
 
-        rb = RollingBenchmark(model_type=self.forecast_model)
+        rb = RollingBenchmark(model_type=self.forecast_model, **self.rb_kwargs)
         task_l = rb.create_rolling_tasks()
 
         # 2.2) create meta dataset for final dataset
@@ -233,13 +276,13 @@ class DDGDA:
         """
         with self._task_path.open("rb") as f:
             tasks = pickle.load(f)
-        rb = RollingBenchmark(rolling_exp="rolling_ds", model_type=self.forecast_model)
+        rb = RollingBenchmark(rolling_exp="rolling_ds", model_type=self.forecast_model, **self.rb_kwargs)
         rb.train_rolling_tasks(tasks)
         rb.ens_rolling()
         rb.update_rolling_rec()
 
     def run_all(self):
-        # 1) file: handler_proxy.pkl
+        # 1) file: handler_proxy.pkl (self.proxy_hd)
         self.dump_data_for_proxy_model()
         # 2)
         # file: internal_data_s20.pkl

examples/benchmarks_dynamic/README.md

@@ -4,15 +4,21 @@ So adapting the forecasting models/strategies to market dynamics is very importa
 
 The table below shows the performances of different solutions on different forecasting models.
 
-## Alpha158 dataset
+## Alpha158 Dataset
+Here is the [crowd sourced version of qlib data](data_collector/crowd_source/README.md): https://github.com/chenditc/investment_data/releases
+```bash
+wget https://github.com/chenditc/investment_data/releases/download/20220720/qlib_bin.tar.gz
+tar -zxvf qlib_bin.tar.gz -C ~/.qlib/qlib_data/cn_data --strip-components=2
+```
 
 | Model Name       | Dataset | IC | ICIR | Rank IC | Rank ICIR | Annualized Return | Information Ratio | Max Drawdown |
 |------------------|---------|----|------|---------|-----------|-------------------|-------------------|--------------|
-| RR[Linear]       |Alpha158 |0.088|0.570|0.102    |0.622      |0.077              |1.175              |-0.086        |
-| DDG-DA[Linear]   |Alpha158 |0.093|0.622|0.106    |0.670      |0.085              |1.213              |-0.093        |
-| RR[LightGBM]     |Alpha158 |0.079|0.566|0.088    |0.592      |0.075              |1.226              |-0.096        |
-| DDG-DA[LightGBM] |Alpha158 |0.084|0.639|0.093    |0.664      |0.099              |1.442              |-0.071        |
+| RR[Linear]       |Alpha158 |0.089|0.577|0.102    |0.627      |0.093              |1.458              |-0.073        |
+| DDG-DA[Linear]   |Alpha158 |0.096|0.636|0.107    |0.677      |0.067              |0.996              |-0.091        |
+| RR[LightGBM]     |Alpha158 |0.082|0.589|0.091    |0.626      |0.077              |1.320              |-0.091        |
+| DDG-DA[LightGBM] |Alpha158 |0.085|0.658|0.094    |0.686      |0.115              |1.792              |-0.068        |
 
 - The label horizon of the `Alpha158` dataset is set to 20.
 - The rolling time intervals are set to 20 trading days.
 - The test rolling periods are from January 2017 to August 2020.
+- The results are based on the crowd-sourced version. The Yahoo version of qlib data does not contain `VWAP`, so all related factors are missing and filled with 0, which leads to a rank-deficient matrix (a matrix does not have full rank) and makes lower-level optimization of DDG-DA can not be solved.

examples/benchmarks_dynamic/baseline/rolling_benchmark.py

@@ -1,13 +1,17 @@
 # Copyright (c) Microsoft Corporation.
 # Licensed under the MIT License.
+from typing import Optional
 from qlib.model.ens.ensemble import RollingEnsemble
 from qlib.utils import init_instance_by_config
 import fire
 import yaml
+import pandas as pd
 from qlib import auto_init
 from pathlib import Path
 from tqdm.auto import tqdm
 from qlib.model.trainer import TrainerR
+from qlib.log import get_module_logger
+from qlib.utils.data import update_config
 from qlib.workflow import R
 from qlib.tests.data import GetData
 
@@ -25,11 +29,40 @@ class RollingBenchmark:
 
     """
 
-    def __init__(self, rolling_exp="rolling_models", model_type="linear") -> None:
+    def __init__(
+        self,
+        rolling_exp: str = "rolling_models",
+        model_type: str = "linear",
+        h_path: Optional[str] = None,
+        train_start: Optional[str] = None,
+        test_end: Optional[str] = None,
+        task_ext_conf: Optional[dict] = None,
+    ) -> None:
+        """
+        Parameters
+        ----------
+        rolling_exp : str
+            The name for the experiments for rolling
+        model_type : str
+            The model to be boosted.
+        h_path : Optional[str]
+            the dumped data handler;
+        test_end : Optional[str]
+            the test end for the data. It is typically used together with the handler
+        train_start : Optional[str]
+            the train start for the data.  It is typically used together with the handler.
+        task_ext_conf : Optional[dict]
+            some option to update the
+        """
         self.step = 20
         self.horizon = 20
         self.rolling_exp = rolling_exp
         self.model_type = model_type
+        self.h_path = h_path
+        self.train_start = train_start
+        self.test_end = test_end
+        self.logger = get_module_logger("RollingBenchmark")
+        self.task_ext_conf = task_ext_conf
 
     def basic_task(self):
         """For fast training rolling"""
@@ -42,6 +75,10 @@ class RollingBenchmark:
             h_path = DIRNAME / "linear_alpha158_handler_horizon{}.pkl".format(self.horizon)
         else:
             raise AssertionError("Model type is not supported!")
+
+        if self.h_path is not None:
+            h_path = Path(self.h_path)
+
         with conf_path.open("r") as f:
             conf = yaml.safe_load(f)
 
@@ -52,6 +89,9 @@ class RollingBenchmark:
 
         task = conf["task"]
 
+        if self.task_ext_conf is not None:
+            task = update_config(task, self.task_ext_conf)
+
         if not h_path.exists():
             h_conf = task["dataset"]["kwargs"]["handler"]
             h = init_instance_by_config(h_conf)
@@ -59,6 +99,15 @@ class RollingBenchmark:
 
         task["dataset"]["kwargs"]["handler"] = f"file://{h_path}"
         task["record"] = ["qlib.workflow.record_temp.SignalRecord"]
+
+        if self.train_start is not None:
+            seg = task["dataset"]["kwargs"]["segments"]["train"]
+            task["dataset"]["kwargs"]["segments"]["train"] = pd.Timestamp(self.train_start), seg[1]
+
+        if self.test_end is not None:
+            seg = task["dataset"]["kwargs"]["segments"]["test"]
+            task["dataset"]["kwargs"]["segments"]["test"] = seg[0], pd.Timestamp(self.test_end)
+        self.logger.info(task)
         return task
 
     def create_rolling_tasks(self):
@@ -93,7 +142,7 @@ class RollingBenchmark:
         """
         Evaluate the combined rolling results
         """
-        for rid, rec in R.list_recorders(experiment_name=self.COMB_EXP).items():
+        for _, rec in R.list_recorders(experiment_name=self.COMB_EXP).items():
             for rt_cls in SigAnaRecord, PortAnaRecord:
                 rt = rt_cls(recorder=rec, skip_existing=True)
                 rt.generate()

examples/rl_order_execution/README.md

@@ -0,0 +1,100 @@
+# RL Example for Order Execution
+
+This folder comprises an example of Reinforcement Learning (RL) workflows for order execution scenario, including both training workflows and backtest workflows.
+
+## Data Processing
+
+### Get Data
+
+```
+python -m qlib.run.get_data qlib_data qlib_data --target_dir ./data/bin --region hs300 --interval 5min
+```
+
+### Generate Pickle-Style Data
+
+To run codes in this example, we need data in pickle format. To achieve this, run following commands (might need a few minutes to finish):
+
+[//]: # (TODO: Instead of dumping dataframe with different format (like `_gen_dataset` and `_gen_day_dataset` in `qlib/contrib/data/highfreq_provider.py`), we encourage to implement different subclass of `Dataset` and `DataHandler`. This will keep the workflow cleaner and interfaces more consistent, and move all the complexity to the subclass.)
+
+```
+python scripts/gen_pickle_data.py -c scripts/pickle_data_config.yml
+python scripts/gen_training_orders.py
+python scripts/merge_orders.py
+```
+
+When finished, the structure under `data/` should be:
+
+```
+data
+├── bin
+├── orders
+└── pickle
+```
+
+## Training
+
+Each training task is specified by a config file. The config file for task `TASKNAME` is `exp_configs/train_TASKNAME.yml`. This example provides two training tasks:
+
+- **PPO**: Method proposed by IJCAL 2020 paper "[An End-to-End Optimal Trade Execution Framework based on Proximal Policy Optimization](https://www.ijcai.org/proceedings/2020/0627.pdf)".
+- **OPDS**: Method proposed by AAAI 2021 paper "[Universal Trading for Order Execution with Oracle Policy Distillation](https://arxiv.org/abs/2103.10860)".
+
+The main differece between these two methods is their reward functions. Please see their config files for details.
+
+Take OPDS as an example, to run the training workflow, run:
+
+```
+python -m qlib.rl.contrib.train_onpolicy --config_path exp_configs/train_opds.yml --run_backtest
+```
+
+Metrics, logs, and checkpoints will be stored under `outputs/opds` (configured by `exp_configs/train_opds.yml`). 
+
+## Backtest
+
+Once the training workflow has completed, the trained model can be used for the backtesting workflow. Still taking OPDS as an example, once training is finished, the latest checkpoint of the model can be found at `outputs/opds/checkpoints/latest.pth`. To run backtest workflow:
+
+1. Uncomment the `weight_file` parameter in `exp_configs/train_opds.yml` (it is commented by default). While it is possible to run the backtesting workflow without setting a checkpoint, this will lead to randomly initialized model results, thus making them meaningless.
+2. Run `python -m qlib.rl.contrib.backtest --config_path exp_configs/backtest_opds.yml`.
+
+The backtest result is stored in `outputs/checkpoints/backtest_result.csv`.
+
+In addition to OPDS and PPO, we also provide TWAP ([Time-weighted average price](https://en.wikipedia.org/wiki/Time-weighted_average_price)) as a weak baseline. The config file for TWAP is `exp_configs/backtest_twap.yml`.
+
+### Gap between backtest and training pipeline's testing
+
+It is worthy to notice that the results of the backtesting process may differ from the results of the testing process used during training.
+This is because different simulators are used to simulate market conditions during training and backtesting.
+In training pipeline, the simplified simulator called `SingleAssetOrderExecutionSimple` is used for efficiency reasons. 
+`SingleAssetOrderExecutionSimple` makes no restriction to trading amounts. 
+No matter what the amount of the order is, it can be completely executed.
+However, during backtesting, a more realistic simulator called `SingleAssetOrderExecution` is used. 
+It takes into account practical constraints in more real-world scenarios (for example, the trading volume must be a multiple of the smallest trading unit).
+As a result, the amount of an order that is actually executed during backtesting may differ from the amount expected to be executed.
+
+If you would like to obtain results that are exactly the same as those obtained during testing in the training pipeline, you could run training pipeline with only backtest phrase.
+In order to do this:
+- Modify the training config. Add the path of the checkpoint you want to use (see following for an example).
+- Run `python -m qlib.rl.contrib.train_onpolicy --config_path PATH/TO/CONFIG --run_backtest --no_training`
+
+```yaml
+...
+policy:
+  class: PPO  # PPO, DQN
+  kwargs:
+    lr: 0.0001
+    weight_file: PATH/TO/CHECKPOINT
+  module_path: qlib.rl.order_execution.policy
+...
+```
+
+## Benchmarks (TBD)
+
+To accurately evaluate the performance of models using Reinforcement Learning algorithms, it's best to run experiments multiple times and compute the average performance across all trials. However, given the time-consuming nature of model training, this is not always feasible. An alternative approach is to run each training task only once, selecting the 10 checkpoints with the highest validation performance to simulate multiple trials. In this example, we use "Price Advantage (PA)" as the metric for selecting these checkpoints. The average performance of these 10 checkpoints on the testing set is as follows:
+
+| **Model**                   | **PA mean with std.** |
+|-----------------------------|-----------------------|
+| OPDS (with PPO policy)      |  0.4785 ± 0.7815      |
+| OPDS (with DQN policy)      | -0.0114 ± 0.5780      |
+| PPO                         | -1.0935 ± 0.0922      |
+| TWAP                        |   ≈ 0.0 ± 0.0         |
+
+The table above also includes TWAP as a rule-based baseline. The ideal PA of TWAP should be 0.0, however, in this example, the order execution is divided into two steps: first, the order is split equally among each half hour, and then each five minutes within each half hour. Since trading is forbidden during the last five minutes of the day, this approach may slightly differ from traditional TWAP over the course of a full day (as there are 5 minutes missing in the last "half hour"). Therefore, the PA of TWAP can be considered as a number that is close to 0.0. To verify this, you may run a TWAP backtest and check the results.

examples/rl_order_execution/exp_configs/backtest_opds.yml

@@ -0,0 +1,53 @@
+order_file: ./data/orders/test_orders.pkl
+start_time: "9:30"
+end_time: "14:54"
+data_granularity: "5min"
+qlib:
+  provider_uri_5min: ./data/bin/
+exchange:
+  limit_threshold: null
+  deal_price: ["$close", "$close"]
+  volume_threshold: null
+strategies:
+  1day:
+    class: SAOEIntStrategy
+    kwargs:
+      data_granularity: 5
+      action_interpreter:
+        class: CategoricalActionInterpreter
+        kwargs:
+          max_step: 8
+          values: 4
+        module_path: qlib.rl.order_execution.interpreter
+      network:
+        class: Recurrent
+        kwargs: {}
+        module_path: qlib.rl.order_execution.network
+      policy:
+        class: PPO  # PPO, DQN
+        kwargs:
+          lr: 0.0001
+          # Restore `weight_file` once the training workflow finishes. You can change the checkpoint file you want to use.
+          # weight_file: outputs/opds/checkpoints/latest.pth
+        module_path: qlib.rl.order_execution.policy
+      state_interpreter:
+        class: FullHistoryStateInterpreter
+        kwargs:
+          data_dim: 5
+          data_ticks: 48
+          max_step: 8
+          processed_data_provider:
+            class: HandlerProcessedDataProvider
+            kwargs:
+              data_dir: ./data/pickle/
+              feature_columns_today: ["$high", "$low", "$open", "$close", "$volume"]
+              feature_columns_yesterday: ["$high_1", "$low_1", "$open_1", "$close_1", "$volume_1"]
+            module_path: qlib.rl.data.native
+        module_path: qlib.rl.order_execution.interpreter
+    module_path: qlib.rl.order_execution.strategy
+  30min:
+    class: TWAPStrategy
+    kwargs: {}
+    module_path: qlib.contrib.strategy.rule_strategy
+concurrency: 16
+output_dir: outputs/opds/

examples/rl_order_execution/exp_configs/backtest_ppo.yml

@@ -0,0 +1,53 @@
+order_file: ./data/orders/test_orders.pkl
+start_time: "9:30"
+end_time: "14:54"
+data_granularity: "5min"
+qlib:
+  provider_uri_5min: ./data/bin/
+exchange:
+  limit_threshold: null
+  deal_price: ["$close", "$close"]
+  volume_threshold: null
+strategies:
+  1day:
+    class: SAOEIntStrategy
+    kwargs:
+      data_granularity: 5
+      action_interpreter:
+        class: CategoricalActionInterpreter
+        kwargs:
+          max_step: 8
+          values: 4
+        module_path: qlib.rl.order_execution.interpreter
+      network:
+        class: Recurrent
+        kwargs: {}
+        module_path: qlib.rl.order_execution.network
+      policy:
+        class: PPO  # PPO, DQN
+        kwargs:
+          lr: 0.0001
+          # Restore `weight_file` once the training workflow finishes. You can change the checkpoint file you want to use.
+          # weight_file: outputs/ppo/checkpoints/latest.pth
+        module_path: qlib.rl.order_execution.policy
+      state_interpreter:
+        class: FullHistoryStateInterpreter
+        kwargs:
+          data_dim: 5
+          data_ticks: 48
+          max_step: 8
+          processed_data_provider:
+            class: HandlerProcessedDataProvider
+            kwargs:
+              data_dir: ./data/pickle/
+              feature_columns_today: ["$high", "$low", "$open", "$close", "$volume"]
+              feature_columns_yesterday: ["$high_1", "$low_1", "$open_1", "$close_1", "$volume_1"]
+            module_path: qlib.rl.data.native
+        module_path: qlib.rl.order_execution.interpreter
+    module_path: qlib.rl.order_execution.strategy
+  30min:
+    class: TWAPStrategy
+    kwargs: {}
+    module_path: qlib.contrib.strategy.rule_strategy
+concurrency: 16
+output_dir: outputs/ppo/

examples/rl_order_execution/exp_configs/backtest_twap.yml

@@ -0,0 +1,21 @@
+order_file: ./data/orders/test_orders.pkl
+start_time: "9:30"
+end_time: "14:54"
+data_granularity: "5min"
+qlib:
+  provider_uri_5min: ./data/bin/
+exchange:
+  limit_threshold: null
+  deal_price: ["$close", "$close"]
+  volume_threshold: null
+strategies:
+  1day:
+    class: TWAPStrategy
+    kwargs: {}
+    module_path: qlib.contrib.strategy.rule_strategy
+  30min:
+    class: TWAPStrategy
+    kwargs: {}
+    module_path: qlib.contrib.strategy.rule_strategy
+concurrency: 16
+output_dir: outputs/twap/

examples/rl_order_execution/exp_configs/train_opds.yml

@@ -0,0 +1,66 @@
+simulator:
+  data_granularity: 5
+  time_per_step: 30
+  vol_limit: null
+env:
+  concurrency: 32
+  parallel_mode: dummy
+action_interpreter:
+  class: CategoricalActionInterpreter
+  kwargs:
+    values: 4
+    max_step: 8
+  module_path: qlib.rl.order_execution.interpreter
+state_interpreter:
+  class: FullHistoryStateInterpreter
+  kwargs:
+    data_dim: 5
+    data_ticks: 48  # 48 = 240 min / 5 min
+    max_step: 8
+    processed_data_provider:
+      class: HandlerProcessedDataProvider
+      kwargs:
+        data_dir: ./data/pickle/
+        feature_columns_today: ["$high", "$low", "$open", "$close", "$volume"]
+        feature_columns_yesterday: ["$high_1", "$low_1", "$open_1", "$close_1", "$volume_1"]
+        backtest: false
+      module_path: qlib.rl.data.native
+  module_path: qlib.rl.order_execution.interpreter
+reward:
+  class: PAPenaltyReward
+  kwargs:
+    penalty: 4.0
+    scale: 0.01
+  module_path: qlib.rl.order_execution.reward
+data:
+  source:
+    order_dir: ./data/orders
+    feature_root_dir: ./data/pickle/
+    feature_columns_today: ["$close0", "$volume0"]
+    feature_columns_yesterday: []
+    total_time: 240
+    default_start_time_index: 0
+    default_end_time_index: 235
+    proc_data_dim: 5
+  num_workers: 0
+  queue_size: 20
+network:
+  class: Recurrent
+  module_path: qlib.rl.order_execution.network
+policy:
+  class: PPO  # PPO, DQN
+  kwargs:
+    lr: 0.0001
+  module_path: qlib.rl.order_execution.policy
+runtime:
+  seed: 42
+  use_cuda: false
+trainer:
+  max_epoch: 500
+  repeat_per_collect: 25
+  earlystop_patience: 50
+  episode_per_collect: 10000
+  batch_size: 1024
+  val_every_n_epoch: 4
+  checkpoint_path: ./outputs/opds
+  checkpoint_every_n_iters: 1

examples/rl_order_execution/exp_configs/train_ppo.yml

@@ -0,0 +1,67 @@
+simulator:
+  data_granularity: 5
+  time_per_step: 30
+  vol_limit: null
+env:
+  concurrency: 32
+  parallel_mode: dummy
+action_interpreter:
+  class: CategoricalActionInterpreter
+  kwargs:
+    values: 4
+    max_step: 8
+  module_path: qlib.rl.order_execution.interpreter
+state_interpreter:
+  class: FullHistoryStateInterpreter
+  kwargs:
+    data_dim: 5
+    data_ticks: 48  # 48 = 240 min / 5 min
+    max_step: 8
+    processed_data_provider:
+      class: HandlerProcessedDataProvider
+      kwargs:
+        data_dir: ./data/pickle/
+        feature_columns_today: ["$high", "$low", "$open", "$close", "$volume"]
+        feature_columns_yesterday: ["$high_1", "$low_1", "$open_1", "$close_1", "$volume_1"]
+        backtest: false
+      module_path: qlib.rl.data.native
+  module_path: qlib.rl.order_execution.interpreter
+reward:
+  class: PPOReward
+  kwargs:
+    max_step: 8
+    start_time_index: 0
+    end_time_index: 46  # 46 = (240 - 5) min / 5 min - 1
+  module_path: qlib.rl.order_execution.reward
+data:
+  source:
+    order_dir: ./data/orders
+    feature_root_dir: ./data/pickle/
+    feature_columns_today: ["$close0", "$volume0"]
+    feature_columns_yesterday: []
+    total_time: 240
+    default_start_time_index: 0
+    default_end_time_index: 235
+    proc_data_dim: 5
+  num_workers: 0
+  queue_size: 20
+network:
+  class: Recurrent
+  module_path: qlib.rl.order_execution.network
+policy:
+  class: PPO  # PPO, DQN
+  kwargs:
+    lr: 0.0001
+  module_path: qlib.rl.order_execution.policy
+runtime:
+  seed: 42
+  use_cuda: false
+trainer:
+  max_epoch: 500
+  repeat_per_collect: 25
+  earlystop_patience: 50
+  episode_per_collect: 10000
+  batch_size: 1024
+  val_every_n_epoch: 4
+  checkpoint_path: ./outputs/ppo
+  checkpoint_every_n_iters: 1

examples/rl_order_execution/scripts/gen_pickle_data.py

@@ -0,0 +1,46 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+import yaml
+import argparse
+import os
+import shutil
+from copy import deepcopy
+
+from qlib.contrib.data.highfreq_provider import HighFreqProvider
+
+loader = yaml.FullLoader
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-c", "--config", type=str, default="config.yml")
+    parser.add_argument("-d", "--dest", type=str, default=".")
+    parser.add_argument("-s", "--split", type=str, choices=["none", "date", "stock", "both"], default="stock")
+    args = parser.parse_args()
+
+    conf = yaml.load(open(args.config), Loader=loader)
+
+    for k, v in conf.items():
+        if isinstance(v, dict) and "path" in v:
+            v["path"] = os.path.join(args.dest, v["path"])
+    provider = HighFreqProvider(**conf)
+
+    # Gen dataframe
+    if "feature_conf" in conf:
+        feature = provider._gen_dataframe(deepcopy(provider.feature_conf))
+    if "backtest_conf" in conf:
+        backtest = provider._gen_dataframe(deepcopy(provider.backtest_conf))
+
+    provider.feature_conf["path"] = os.path.splitext(provider.feature_conf["path"])[0] + "/"
+    provider.backtest_conf["path"] = os.path.splitext(provider.backtest_conf["path"])[0] + "/"
+    # Split by date
+    if args.split == "date" or args.split == "both":
+        provider._gen_day_dataset(deepcopy(provider.feature_conf), "feature")
+        provider._gen_day_dataset(deepcopy(provider.backtest_conf), "backtest")
+
+    # Split by stock
+    if args.split == "stock" or args.split == "both":
+        provider._gen_stock_dataset(deepcopy(provider.feature_conf), "feature")
+        provider._gen_stock_dataset(deepcopy(provider.backtest_conf), "backtest")
+
+    shutil.rmtree("stat/", ignore_errors=True)

examples/rl_order_execution/scripts/gen_training_orders.py

@@ -0,0 +1,53 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+import os
+import numpy as np
+import pandas as pd
+
+from pathlib import Path
+
+DATA_PATH = Path(os.path.join("data", "pickle", "backtest"))
+OUTPUT_PATH = Path(os.path.join("data", "orders"))
+
+
+def generate_order(stock: str, start_idx: int, end_idx: int) -> bool:
+    dataset = pd.read_pickle(DATA_PATH / f"{stock}.pkl")
+    df = dataset.handler.fetch(level=None).reset_index()
+    if len(df) == 0 or df.isnull().values.any() or min(df["$volume0"]) < 1e-5:
+        return False
+
+    df["date"] = df["datetime"].dt.date.astype("datetime64")
+    df = df.set_index(["instrument", "datetime", "date"])
+    df = df.groupby("date").take(range(start_idx, end_idx)).droplevel(level=0)
+
+    order_all = pd.DataFrame(df.groupby(level=(2, 0)).mean().dropna())
+    order_all["amount"] = np.random.lognormal(-3.28, 1.14) * order_all["$volume0"]
+    order_all = order_all[order_all["amount"] > 0.0]
+    order_all["order_type"] = 0
+    order_all = order_all.drop(columns=["$volume0"])
+
+    order_train = order_all[order_all.index.get_level_values(0) <= pd.Timestamp("2021-06-30")]
+    order_test = order_all[order_all.index.get_level_values(0) > pd.Timestamp("2021-06-30")]
+    order_valid = order_test[order_test.index.get_level_values(0) <= pd.Timestamp("2021-09-30")]
+    order_test = order_test[order_test.index.get_level_values(0) > pd.Timestamp("2021-09-30")]
+
+    for order, tag in zip((order_train, order_valid, order_test, order_all), ("train", "valid", "test", "all")):
+        path = OUTPUT_PATH / tag
+        os.makedirs(path, exist_ok=True)
+        if len(order) > 0:
+            order.to_pickle(path / f"{stock}.pkl.target")
+    return True
+
+
+np.random.seed(1234)
+file_list = sorted(os.listdir(DATA_PATH))
+stocks = [f.replace(".pkl", "") for f in file_list]
+np.random.shuffle(stocks)
+
+cnt = 0
+for stock in stocks:
+    if generate_order(stock, 0, 240 // 5 - 1):
+        cnt += 1
+        if cnt == 100:
+            break

examples/rl_order_execution/scripts/merge_orders.py

@@ -0,0 +1,15 @@
+import pickle
+import os
+import pandas as pd
+from tqdm import tqdm
+
+for tag in ["test", "valid"]:
+    files = os.listdir(os.path.join("data/orders/", tag))
+    dfs = []
+    for f in tqdm(files):
+        df = pickle.load(open(os.path.join("data/orders/", tag, f), "rb"))
+        df = df.drop(["$close0"], axis=1)
+        dfs.append(df)
+
+    total_df = pd.concat(dfs)
+    pickle.dump(total_df, open(os.path.join("data", "orders", f"{tag}_orders.pkl"), "wb"))

examples/rl_order_execution/scripts/pickle_data_config.yml

@@ -0,0 +1,77 @@
+# start & end time for training/validation/test datasets
+start_time: !!str &start 2020-01-01
+end_time: !!str &end 2021-12-31
+train_end_time: !!str &tend 2021-06-30
+valid_start_time: !!str &vstart 2021-07-01
+valid_end_time: !!str &vend 2021-09-30
+test_start_time: !!str &tstart 2021-10-01
+# the instrument set
+instruments: &ins csi300s19_22
+# qlib related configuration
+qlib_conf:
+    provider_uri: 
+        5min: ./data/bin # path to generated qlib bin
+    redis_port: 233
+feature_conf:
+    path: ./data/pickle/feature.pkl # output path of feature
+    class: DatasetH
+    module_path: qlib.data.dataset
+    kwargs:
+        handler:
+            class: HighFreqGeneralHandler
+            module_path: qlib.contrib.data.highfreq_handler
+            kwargs:
+                start_time: *start
+                end_time: *end
+                fit_start_time: *start
+                fit_end_time: *tend
+                instruments: *ins
+                day_length: 240 # how many minutes in one trading day
+                freq: 5min
+                columns: ["$open", "$high", "$low", "$close"]
+                infer_processors:
+                - class: HighFreqNorm
+                  module_path: qlib.contrib.data.highfreq_processor
+                  kwargs:
+                    feature_save_dir: ./stat/  #  output path of statistics of features (for feature normalization)
+                    norm_groups: 
+                        price: 8
+                        volume: 2
+                inst_processors:
+                - class: TimeRangeFlt
+                  module_path: qlib.data.dataset.processor
+                  kwargs:
+                    start_time: "2020-01-01"
+                    end_time: "2021-12-31"
+                    freq: 5min
+        segments:
+            train: !!python/tuple [*start, *tend]
+            valid: !!python/tuple [*vstart, *vend]
+            test: !!python/tuple [*tstart, *end]
+backtest_conf:
+    path: ./data/pickle/backtest.pkl # output path of backtest
+    class: DatasetH
+    module_path: qlib.data.dataset
+    kwargs:
+        handler:
+            class: HighFreqGeneralBacktestHandler
+            module_path: qlib.contrib.data.highfreq_handler
+            kwargs:
+                start_time: *start
+                end_time: *end
+                instruments: *ins
+                day_length: 240
+                freq: 5min
+                columns: ["$close", "$volume"]
+                inst_processors:
+                - class: TimeRangeFlt
+                  module_path: qlib.data.dataset.processor
+                  kwargs:
+                    start_time: "2020-01-01"
+                    end_time: "2021-12-31"
+                    freq: 5min
+        segments:
+            train: !!python/tuple [*start, *tend]
+            valid: !!python/tuple [*vstart, *vend]
+            test: !!python/tuple [*tstart, *end]
+freq: 5min

examples/run_all_model.py

@@ -117,8 +117,10 @@ def get_all_folders(models, exclude) -> dict:
 
 
 # function to get all the files under the model folder
-def get_all_files(folder_path, dataset) -> (str, str):
-    yaml_path = str(Path(f"{folder_path}") / f"*{dataset}*.yaml")
+def get_all_files(folder_path, dataset, universe="") -> (str, str):
+    if universe != "":
+        universe = f"_{universe}"
+    yaml_path = str(Path(f"{folder_path}") / f"*{dataset}{universe}.yaml")
     req_path = str(Path(f"{folder_path}") / f"*.txt")
     yaml_file = glob.glob(yaml_path)
     req_file = glob.glob(req_path)
@@ -224,6 +226,7 @@ class ModelRunner:
         times=1,
         models=None,
         dataset="Alpha360",
+        universe="",
         exclude=False,
         qlib_uri: str = "git+https://github.com/microsoft/qlib#egg=pyqlib",
         exp_folder_name: str = "run_all_model_records",
@@ -245,9 +248,12 @@ class ModelRunner:
             determines whether the model being used is excluded or included.
         dataset : str
             determines the dataset to be used for each model.
+        universe  : str
+            the stock universe of the dataset.
+            default "" indicates that
         qlib_uri : str
             the uri to install qlib with pip
-            it could be url on the we or local path (NOTE: the local path must be a absolute path)
+            it could be URI on the remote or local path (NOTE: the local path must be an absolute path)
         exp_folder_name: str
             the name of the experiment folder
         wait_before_rm_env : bool
@@ -259,6 +265,15 @@ class ModelRunner:
         -------
         Here are some use cases of the function in the bash:
 
+        The run_all_models  will decide which config to run based no `models` `dataset`  `universe`
+        Example 1):
+
+            models="lightgbm", dataset="Alpha158", universe="" will result in running the following config
+            examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158.yaml
+
+            models="lightgbm", dataset="Alpha158", universe="csi500" will result in running the following config
+            examples/benchmarks/LightGBM/workflow_config_lightgbm_Alpha158_csi500.yaml
+
         .. code-block:: bash
 
             # Case 1 - run all models multiple times
@@ -279,6 +294,9 @@ class ModelRunner:
             # Case 6 - run other models except those are given as arguments for one time
             python run_all_model.py run --models=[mlp,tft,sfm] --exclude=True
 
+            # Case 7 - run lightgbm model on csi500.
+            python run_all_model.py run 3 lightgbm Alpha158 csi500
+
         """
         self._init_qlib(exp_folder_name)
 
@@ -290,7 +308,7 @@ class ModelRunner:
         for fn in folders:
             # get all files
             sys.stderr.write("Retrieving files...\n")
-            yaml_path, req_path = get_all_files(folders[fn], dataset)
+            yaml_path, req_path = get_all_files(folders[fn], dataset, universe=universe)
             if yaml_path is None:
                 sys.stderr.write(f"There is no {dataset}.yaml file in {folders[fn]}")
                 continue

examples/tutorial/detailed_workflow.ipynb

@@ -88,6 +88,7 @@
    "outputs": [],
    "source": [
     "from qlib.tests.data import GetData\n",
+    "\n",
     "GetData().qlib_data(exists_skip=True)"
    ]
   },
@@ -99,6 +100,7 @@
    "outputs": [],
    "source": [
     "import qlib\n",
+    "\n",
     "qlib.init()"
    ]
   },
@@ -134,7 +136,8 @@
    "outputs": [],
    "source": [
     "from qlib.data import D\n",
-    "D.calendar(start_time='2010-01-01', end_time='2017-12-31', freq='day')[:2]  # calendar data"
+    "\n",
+    "print(D.calendar(start_time=\"2010-01-01\", end_time=\"2017-12-31\", freq=\"day\")[:2])  # calendar data"
    ]
   },
   {
@@ -152,7 +155,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "df = D.features(['SH601216'], ['$open', '$high', '$low', '$close', '$factor'], start_time='2020-05-01', end_time='2020-05-31')   "
+    "df = D.features(\n",
+    "    [\"SH601216\"],\n",
+    "    [\"$open\", \"$high\", \"$low\", \"$close\", \"$factor\"],\n",
+    "    start_time=\"2020-05-01\",\n",
+    "    end_time=\"2020-05-31\",\n",
+    ")"
    ]
   },
   {
@@ -163,11 +171,18 @@
    "outputs": [],
    "source": [
     "import plotly.graph_objects as go\n",
-    "fig = go.Figure(data=[go.Candlestick(x=df.index.get_level_values(\"datetime\"),\n",
-    "                open=df['$open'],\n",
-    "                high=df['$high'],\n",
-    "                low=df['$low'],\n",
-    "                close=df['$close'])])\n",
+    "\n",
+    "fig = go.Figure(\n",
+    "    data=[\n",
+    "        go.Candlestick(\n",
+    "            x=df.index.get_level_values(\"datetime\"),\n",
+    "            open=df[\"$open\"],\n",
+    "            high=df[\"$high\"],\n",
+    "            low=df[\"$low\"],\n",
+    "            close=df[\"$close\"],\n",
+    "        )\n",
+    "    ]\n",
+    ")\n",
     "fig.show()"
    ]
   },
@@ -197,11 +212,18 @@
    "outputs": [],
    "source": [
     "import plotly.graph_objects as go\n",
-    "fig = go.Figure(data=[go.Candlestick(x=df.index.get_level_values(\"datetime\"),\n",
-    "                open=df['$open'] / df['$factor'],\n",
-    "                high=df['$high'] / df['$factor'],\n",
-    "                low=df['$low'] / df['$factor'],\n",
-    "                close=df['$close'] / df['$factor'])])\n",
+    "\n",
+    "fig = go.Figure(\n",
+    "    data=[\n",
+    "        go.Candlestick(\n",
+    "            x=df.index.get_level_values(\"datetime\"),\n",
+    "            open=df[\"$open\"] / df[\"$factor\"],\n",
+    "            high=df[\"$high\"] / df[\"$factor\"],\n",
+    "            low=df[\"$low\"] / df[\"$factor\"],\n",
+    "            close=df[\"$close\"] / df[\"$factor\"],\n",
+    "        )\n",
+    "    ]\n",
+    ")\n",
     "fig.show()"
    ]
   },
@@ -240,7 +262,7 @@
    "outputs": [],
    "source": [
     "# dynamic universe\n",
-    "universe = D.list_instruments(D.instruments('csi100'),  start_time='2010-01-01', end_time='2020-12-31')\n",
+    "universe = D.list_instruments(D.instruments(\"csi100\"), start_time=\"2010-01-01\", end_time=\"2020-12-31\")\n",
     "pprint(universe)"
    ]
   },
@@ -271,8 +293,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "df = D.features(D.instruments('csi100'), ['$close'], start_time='2010-01-01', end_time='2020-12-31')   \n",
-    "df.groupby('datetime').size().plot()"
+    "df = D.features(D.instruments(\"csi100\"), [\"$close\"], start_time=\"2010-01-01\", end_time=\"2020-12-31\")\n",
+    "df.groupby(\"datetime\").size().plot()"
    ]
   },
   {
@@ -313,8 +335,7 @@
     "    !cd ../../scripts/data_collector/pit/ && pip install -r requirements.txt\n",
     "    !cd ../../scripts/data_collector/pit/ && python collector.py download_data --source_dir ~/.qlib/stock_data/source/pit --start 2000-01-01 --end 2020-01-01 --interval quarterly --symbol_regex \"^(600519|000725).*\"\n",
     "    !cd ../../scripts/data_collector/pit/ && python collector.py normalize_data --interval quarterly --source_dir ~/.qlib/stock_data/source/pit --normalize_dir ~/.qlib/stock_data/source/pit_normalized\n",
-    "    !cd ../../scripts/ && python dump_pit.py dump --csv_path ~/.qlib/stock_data/source/pit_normalized --qlib_dir ~/.qlib/qlib_data/cn_data --interval quarterly\n",
-    "    pass"
+    "    !cd ../../scripts/ && python dump_pit.py dump --csv_path ~/.qlib/stock_data/source/pit_normalized --qlib_dir ~/.qlib/qlib_data/cn_data --interval quarterly"
    ]
   },
   {
@@ -338,7 +359,13 @@
    "outputs": [],
    "source": [
     "instruments = [\"sh600519\"]\n",
-    "data = D.features(instruments, ['P($$roewa_q)'], start_time=\"2019-01-01\", end_time=\"2019-07-19\", freq=\"day\")"
+    "data = D.features(\n",
+    "    instruments,\n",
+    "    [\"P($$roewa_q)\"],\n",
+    "    start_time=\"2019-01-01\",\n",
+    "    end_time=\"2019-07-19\",\n",
+    "    freq=\"day\",\n",
+    ")"
    ]
   },
   {
@@ -366,7 +393,10 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "D.features([\"sh600519\"], ['(EMA($close, 12) - EMA($close, 26))/$close - EMA((EMA($close, 12) - EMA($close, 26))/$close, 9)/$close'])"
+    "D.features(\n",
+    "    [\"sh600519\"],\n",
+    "    [\"(EMA($close, 12) - EMA($close, 26))/$close - EMA((EMA($close, 12) - EMA($close, 26))/$close, 9)/$close\"],\n",
+    ")"
    ]
   },
   {
@@ -418,7 +448,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "qdl = QlibDataLoader(config=(['$close / Ref($close, 10)'], ['RET10']))"
+    "qdl = QlibDataLoader(config=([\"$close / Ref($close, 10)\"], [\"RET10\"]))"
    ]
   },
   {
@@ -428,7 +458,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "qdl.load(instruments=['sh600519'], start_time='20190101', end_time='20191231')"
+    "qdl.load(instruments=[\"sh600519\"], start_time=\"20190101\", end_time=\"20191231\")"
    ]
   },
   {
@@ -456,7 +486,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "df = qdl.load(instruments=['sh600519'], start_time='20190101', end_time='20191231')"
+    "df = qdl.load(instruments=[\"sh600519\"], start_time=\"20190101\", end_time=\"20191231\")"
    ]
   },
   {
@@ -476,7 +506,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "df.plot(kind='hist')"
+    "df.plot(kind=\"hist\")"
    ]
   },
   {
@@ -508,9 +538,16 @@
    "source": [
     "# NOTE: normally, the training & validation time range will be  `fit_start_time` ， `fit_end_time`\n",
     "# however，all the components are decomposed, so the training & validation time range is unknown when preprocessing.\n",
-    "dh = DataHandlerLP(instruments=['sh600519'], start_time='20170101', end_time='20191231',\n",
-    "             infer_processors=[ZScoreNorm(fit_start_time='20170101', fit_end_time='20181231'), Fillna()],\n",
-    "             data_loader=qdl)"
+    "dh = DataHandlerLP(\n",
+    "    instruments=[\"sh600519\"],\n",
+    "    start_time=\"20170101\",\n",
+    "    end_time=\"20191231\",\n",
+    "    infer_processors=[\n",
+    "        ZScoreNorm(fit_start_time=\"20170101\", fit_end_time=\"20181231\"),\n",
+    "        Fillna(),\n",
+    "    ],\n",
+    "    data_loader=qdl,\n",
+    ")"
    ]
   },
   {
@@ -550,7 +587,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "df.plot(kind='hist')"
+    "df.plot(kind=\"hist\")"
    ]
   },
   {
@@ -586,7 +623,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ds = DatasetH(dh, segments={\"train\": ('20180101', '20181231'), \"valid\": ('20190101', '20191231')})"
+    "ds = DatasetH(dh, segments={\"train\": (\"20180101\", \"20181231\"), \"valid\": (\"20190101\", \"20191231\")})"
    ]
   },
   {
@@ -596,7 +633,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ds.prepare('train')"
+    "ds.prepare(\"train\")"
    ]
   },
   {
@@ -606,7 +643,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ds.prepare('valid')"
+    "ds.prepare(\"valid\")"
    ]
   },
   {
@@ -628,8 +665,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ds = TSDatasetH(step_len=10, handler=dh, segments={\"train\": ('20180101', '20181231'), \"valid\": ('20190101', '20191231')})\n",
-    "train_sampler = ds.prepare('train')"
+    "ds = TSDatasetH(\n",
+    "    step_len=10,\n",
+    "    handler=dh,\n",
+    "    segments={\"train\": (\"20180101\", \"20181231\"), \"valid\": (\"20190101\", \"20191231\")},\n",
+    ")\n",
+    "train_sampler = ds.prepare(\"train\")"
    ]
   },
   {
@@ -649,7 +690,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "train_sampler[0] # Retrieving the first example"
+    "train_sampler[0]  # Retrieving the first example"
    ]
   },
   {
@@ -659,7 +700,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "train_sampler['2018-01-08', 'sh600519']  # get the time series by <'timestamp', 'instrument_id'> index"
+    "train_sampler[\"2018-01-08\", \"sh600519\"]  # get the time series by <'timestamp', 'instrument_id'> index"
    ]
   },
   {
@@ -682,11 +723,11 @@
    "outputs": [],
    "source": [
     "handler_kwargs = {\n",
-    "        \"start_time\": \"2008-01-01\",\n",
-    "        \"end_time\": \"2020-08-01\",\n",
-    "        \"fit_start_time\": \"2008-01-01\",\n",
-    "        \"fit_end_time\": \"2014-12-31\",\n",
-    "        \"instruments\": MARKET,\n",
+    "    \"start_time\": \"2008-01-01\",\n",
+    "    \"end_time\": \"2020-08-01\",\n",
+    "    \"fit_start_time\": \"2008-01-01\",\n",
+    "    \"fit_end_time\": \"2014-12-31\",\n",
+    "    \"instruments\": MARKET,\n",
     "}\n",
     "handler_conf = {\n",
     "    \"class\": \"Alpha158\",\n",
@@ -735,6 +776,7 @@
    "outputs": [],
    "source": [
     "from qlib.contrib.data.handler import Alpha158\n",
+    "\n",
     "hd = Alpha158(**handler_kwargs)"
    ]
   },
@@ -826,7 +868,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "hd.process_type # appending type"
+    "hd.process_type  # appending type"
    ]
   },
   {
@@ -857,16 +899,16 @@
    "outputs": [],
    "source": [
     "dataset_conf = {\n",
-    "        \"class\": \"DatasetH\",\n",
-    "        \"module_path\": \"qlib.data.dataset\",\n",
-    "        \"kwargs\": {\n",
-    "            \"handler\": hd,\n",
-    "            \"segments\": {\n",
-    "                \"train\": (\"2008-01-01\", \"2014-12-31\"),\n",
-    "                \"valid\": (\"2015-01-01\", \"2016-12-31\"),\n",
-    "                \"test\": (\"2017-01-01\", \"2020-08-01\"),\n",
-    "            },\n",
+    "    \"class\": \"DatasetH\",\n",
+    "    \"module_path\": \"qlib.data.dataset\",\n",
+    "    \"kwargs\": {\n",
+    "        \"handler\": hd,\n",
+    "        \"segments\": {\n",
+    "            \"train\": (\"2008-01-01\", \"2014-12-31\"),\n",
+    "            \"valid\": (\"2015-01-01\", \"2016-12-31\"),\n",
+    "            \"test\": (\"2017-01-01\", \"2020-08-01\"),\n",
     "        },\n",
+    "    },\n",
     "}"
    ]
   },
@@ -908,7 +950,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = init_instance_by_config({\n",
+    "model = init_instance_by_config(\n",
+    "    {\n",
     "        \"class\": \"LGBModel\",\n",
     "        \"module_path\": \"qlib.contrib.model.gbdt\",\n",
     "        \"kwargs\": {\n",
@@ -922,7 +965,8 @@
     "            \"num_leaves\": 210,\n",
     "            \"num_threads\": 20,\n",
     "        },\n",
-    "})"
+    "    }\n",
+    ")"
    ]
   },
   {
@@ -938,7 +982,7 @@
     "    R.save_objects(trained_model=model)\n",
     "\n",
     "    rec = R.get_recorder()\n",
-    "    rid = rec.id # save the record id\n",
+    "    rid = rec.id  # save the record id\n",
     "\n",
     "    # Inference and saving signal\n",
     "    sr = SignalRecord(model, dataset, rec)\n",
@@ -1001,12 +1045,11 @@
     "\n",
     "# backtest and analysis\n",
     "with R.start(experiment_name=EXP_NAME, recorder_id=rid, resume=True):\n",
-    "\n",
     "    # signal-based analysis\n",
     "    rec = R.get_recorder()\n",
     "    sar = SigAnaRecord(rec)\n",
     "    sar.generate()\n",
-    "    \n",
+    "\n",
     "    #  portfolio-based analysis: backtest\n",
     "    par = PortAnaRecord(rec, port_analysis_config, \"day\")\n",
     "    par.generate()"
@@ -1137,7 +1180,7 @@
    "outputs": [],
    "source": [
     "label_df = dataset.prepare(\"test\", col_set=\"label\")\n",
-    "label_df.columns = ['label']"
+    "label_df.columns = [\"label\"]"
    ]
   },
   {

examples/workflow_by_code.ipynb

@@ -38,6 +38,9 @@
     "    # install qlib\n",
     "    ! pip install --upgrade numpy\n",
     "    ! pip install pyqlib\n",
+    "    if \"google.colab\" in sys.modules:\n",
+    "        # The Google colab environment is a little outdated. We have to downgrade the pyyaml to make it compatible with other packages\n",
+    "        ! pip install pyyaml==5.4.1\n",
     "    # reload\n",
     "    site.main()\n",
     "\n",
@@ -47,7 +50,8 @@
     "    scripts_dir = Path(\"~/tmp/qlib_code/scripts\").expanduser().resolve()\n",
     "    scripts_dir.mkdir(parents=True, exist_ok=True)\n",
     "    import requests\n",
-    "    with requests.get(\"https://raw.githubusercontent.com/microsoft/qlib/main/scripts/get_data.py\") as resp:\n",
+    "\n",
+    "    with requests.get(\"https://raw.githubusercontent.com/microsoft/qlib/main/scripts/get_data.py\", timeout=10) as resp:\n",
     "        with open(scripts_dir.joinpath(\"get_data.py\"), \"wb\") as fp:\n",
     "            fp.write(resp.content)"
    ]
@@ -58,14 +62,13 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "\n",
     "import qlib\n",
     "import pandas as pd\n",
     "from qlib.constant import REG_CN\n",
     "from qlib.utils import exists_qlib_data, init_instance_by_config\n",
     "from qlib.workflow import R\n",
     "from qlib.workflow.record_temp import SignalRecord, PortAnaRecord\n",
-    "from qlib.utils import flatten_dict\n"
+    "from qlib.utils import flatten_dict"
    ]
   },
   {
@@ -83,6 +86,7 @@
     "    print(f\"Qlib data is not found in {provider_uri}\")\n",
     "    sys.path.append(str(scripts_dir))\n",
     "    from get_data import GetData\n",
+    "\n",
     "    GetData().qlib_data(target_dir=provider_uri, region=REG_CN)\n",
     "qlib.init(provider_uri=provider_uri, region=REG_CN)"
    ]
@@ -166,7 +170,7 @@
     "    R.log_params(**flatten_dict(task))\n",
     "    model.fit(dataset)\n",
     "    R.save_objects(trained_model=model)\n",
-    "    rid = R.get_recorder().id\n"
+    "    rid = R.get_recorder().id"
    ]
   },
   {
@@ -235,7 +239,7 @@
     "\n",
     "    # backtest & analysis\n",
     "    par = PortAnaRecord(recorder, port_analysis_config, \"day\")\n",
-    "    par.generate()\n"
+    "    par.generate()"
    ]
   },
   {
@@ -253,6 +257,7 @@
    "source": [
     "from qlib.contrib.report import analysis_model, analysis_position\n",
     "from qlib.data import D\n",
+    "\n",
     "recorder = R.get_recorder(recorder_id=ba_rid, experiment_name=\"backtest_analysis\")\n",
     "print(recorder)\n",
     "pred_df = recorder.load_object(\"pred.pkl\")\n",
@@ -314,7 +319,7 @@
    "outputs": [],
    "source": [
     "label_df = dataset.prepare(\"test\", col_set=\"label\")\n",
-    "label_df.columns = ['label']"
+    "label_df.columns = [\"label\"]"
    ]
   },
   {

examples/workflow_by_code.py

@@ -1,6 +1,12 @@
 #  Copyright (c) Microsoft Corporation.
 #  Licensed under the MIT License.
+"""
+Qlib provides two kinds of interfaces. 
+(1) Users could define the Quant research workflow by a simple configuration.
+(2) Qlib is designed in a modularized way and supports creating research workflow by code just like building blocks.
 
+The interface of (1) is `qrun XXX.yaml`.  The interface of (2) is script like this, which nearly does the same thing as `qrun XXX.yaml`
+"""
 import qlib
 from qlib.constant import REG_CN
 from qlib.utils import init_instance_by_config, flatten_dict

qlib/__init__.py

@@ -2,7 +2,7 @@
 # Licensed under the MIT License.
 from pathlib import Path
 
-__version__ = "0.8.5.99"
+__version__ = "0.9.1.99"
 __version__bak = __version__  # This version is backup for QlibConfig.reset_qlib_version
 import os
 from typing import Union
@@ -34,8 +34,7 @@ def init(default_conf="client", **kwargs):
     from .config import C  # pylint: disable=C0415
     from .data.cache import H  # pylint: disable=C0415
 
-    # FIXME: this logger ignored the level in config
-    logger = get_module_logger("Initialization", level=logging.INFO)
+    logger = get_module_logger("Initialization")
 
     skip_if_reg = kwargs.pop("skip_if_reg", False)
     if skip_if_reg and C.registered:
@@ -48,6 +47,7 @@ def init(default_conf="client", **kwargs):
     if clear_mem_cache:
         H.clear()
     C.set(default_conf, **kwargs)
+    get_module_logger.setLevel(C.logging_level)
 
     # mount nfs
     for _freq, provider_uri in C.provider_uri.items():
@@ -94,7 +94,7 @@ def _mount_nfs_uri(provider_uri, mount_path, auto_mount: bool = False):
     else:
         # Judging system type
         sys_type = platform.system()
-        if "win" in sys_type.lower():
+        if "windows" in sys_type.lower():
             # system: window
             exec_result = os.popen(f"mount -o anon {provider_uri} {mount_path}")
             result = exec_result.read()
@@ -113,6 +113,8 @@ def _mount_nfs_uri(provider_uri, mount_path, auto_mount: bool = False):
             # system: linux/Unix/Mac
             # check mount
             _remote_uri = provider_uri[:-1] if provider_uri.endswith("/") else provider_uri
+            # `mount a /b/c` is different from `mount a /b/c/`. So we convert it into string to make sure handling it accurately
+            mount_path = str(mount_path)
             _mount_path = mount_path[:-1] if mount_path.endswith("/") else mount_path
             _check_level_num = 2
             _is_mount = False

qlib/backtest/__init__.py

@@ -2,8 +2,12 @@
 # Licensed under the MIT License.
 
 from __future__ import annotations
+
 import copy
-from typing import List, Tuple, Union, TYPE_CHECKING
+from pathlib import Path
+from typing import TYPE_CHECKING, Any, Generator, List, Optional, Tuple, Union
+
+import pandas as pd
 
 from .account import Account
 
@@ -11,15 +15,14 @@ if TYPE_CHECKING:
     from ..strategy.base import BaseStrategy
     from .executor import BaseExecutor
     from .decision import BaseTradeDecision
-from .position import Position
-from .exchange import Exchange
-from .backtest import backtest_loop
-from .backtest import collect_data_loop
-from .utils import CommonInfrastructure
-from .decision import Order
-from ..utils import init_instance_by_config
-from ..log import get_module_logger
+
 from ..config import C
+from ..log import get_module_logger
+from ..utils import init_instance_by_config
+from .backtest import INDICATOR_METRIC, PORT_METRIC, backtest_loop, collect_data_loop
+from .decision import Order
+from .exchange import Exchange
+from .utils import CommonInfrastructure
 
 # make import more user-friendly by adding `from qlib.backtest import STH`
 
@@ -28,26 +31,35 @@ logger = get_module_logger("backtest caller")
 
 
 def get_exchange(
-    exchange=None,
-    freq="day",
-    start_time=None,
-    end_time=None,
-    codes="all",
-    subscribe_fields=[],
-    open_cost=0.0015,
-    close_cost=0.0025,
-    min_cost=5.0,
-    limit_threshold=None,
-    deal_price: Union[str, Tuple[str], List[str]] = None,
-    **kwargs,
-):
+    exchange: Union[str, dict, object, Path] = None,
+    freq: str = "day",
+    start_time: Union[pd.Timestamp, str] = None,
+    end_time: Union[pd.Timestamp, str] = None,
+    codes: Union[list, str] = "all",
+    subscribe_fields: list = [],
+    open_cost: float = 0.0015,
+    close_cost: float = 0.0025,
+    min_cost: float = 5.0,
+    limit_threshold: Union[Tuple[str, str], float, None] | None = None,
+    deal_price: Union[str, Tuple[str, str], List[str]] | None = None,
+    **kwargs: Any,
+) -> Exchange:
     """get_exchange
 
     Parameters
     ----------
 
     # exchange related arguments
-    exchange: Exchange().
+    exchange: Exchange
+        It could be None or any types that are acceptable by `init_instance_by_config`.
+    freq: str
+        frequency of data.
+    start_time: Union[pd.Timestamp, str]
+        closed start time for backtest.
+    end_time: Union[pd.Timestamp, str]
+        closed end time for backtest.
+    codes: Union[list, str]
+        list stock_id list or a string of instruments (i.e. all, csi500, sse50)
     subscribe_fields: list
         subscribe fields.
     open_cost : float
@@ -57,12 +69,10 @@ def get_exchange(
     min_cost : float
         min transaction cost.  It is an absolute amount of cost instead of a ratio of your order's deal amount.
         e.g. You must pay at least 5 yuan of commission regardless of your order's deal amount.
-    trade_unit : int
-        Included in kwargs.  Please refer to the docs of `__init__` of `Exchange`
-    deal_price: Union[str, Tuple[str], List[str]]
+    deal_price: Union[str, Tuple[str, str], List[str]]
                 The `deal_price` supports following two types of input
                 - <deal_price> : str
-                - (<buy_price>, <sell_price>): Tuple[str] or List[str]
+                - (<buy_price>, <sell_price>): Tuple[str, str] or List[str]
 
                 <deal_price>, <buy_price> or <sell_price> := <price>
                 <price> := str
@@ -101,10 +111,14 @@ def get_exchange(
 
 
 def create_account_instance(
-    start_time, end_time, benchmark: str, account: Union[float, int, dict], pos_type: str = "Position"
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    benchmark: Optional[str],
+    account: Union[float, int, dict],
+    pos_type: str = "Position",
 ) -> Account:
     """
-    # TODO: is very strange pass benchmark_config in the account(maybe for report)
+    # TODO: is very strange pass benchmark_config in the account (maybe for report)
     # There should be a post-step to process the report.
 
     Parameters
@@ -132,42 +146,42 @@ def create_account_instance(
             key "cash" means initial cash.
             key "stock1" means the information of first stock with amount and price(optional).
             ...
+    pos_type: str
+        Postion type.
     """
     if isinstance(account, (int, float)):
-        pos_kwargs = {"init_cash": account}
+        init_cash = account
+        position_dict = {}
     elif isinstance(account, dict):
-        init_cash = account["cash"]
-        del account["cash"]
-        pos_kwargs = {
-            "init_cash": init_cash,
-            "position_dict": account,
-        }
+        init_cash = account.pop("cash")
+        position_dict = account
     else:
-        raise ValueError("account must be in (int, float, Position)")
+        raise ValueError("account must be in (int, float, dict)")
 
-    kwargs = {
-        "init_cash": account,
-        "benchmark_config": {
+    return Account(
+        init_cash=init_cash,
+        position_dict=position_dict,
+        pos_type=pos_type,
+        benchmark_config={}
+        if benchmark is None
+        else {
             "benchmark": benchmark,
             "start_time": start_time,
             "end_time": end_time,
         },
-        "pos_type": pos_type,
-    }
-    kwargs.update(pos_kwargs)
-    return Account(**kwargs)
+    )
 
 
 def get_strategy_executor(
-    start_time,
-    end_time,
-    strategy: BaseStrategy,
-    executor: BaseExecutor,
-    benchmark: str = "SH000300",
-    account: Union[float, int, Position] = 1e9,
-    exchange_kwargs: dict = {},
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    strategy: Union[str, dict, object, Path],
+    executor: Union[str, dict, object, Path],
+    benchmark: Optional[str] = "SH000300",
+    account: Union[float, int, dict] = 1e9,
+    exchange_kwargs: Union[dict, Exchange] = {},  # TODO: rename parameter
     pos_type: str = "Position",
-):
+) -> Tuple[BaseStrategy, BaseExecutor]:
 
     # NOTE:
     # - for avoiding recursive import
@@ -176,15 +190,22 @@ def get_strategy_executor(
     from .executor import BaseExecutor  # pylint: disable=C0415
 
     trade_account = create_account_instance(
-        start_time=start_time, end_time=end_time, benchmark=benchmark, account=account, pos_type=pos_type
+        start_time=start_time,
+        end_time=end_time,
+        benchmark=benchmark,
+        account=account,
+        pos_type=pos_type,
     )
 
-    exchange_kwargs = copy.copy(exchange_kwargs)
-    if "start_time" not in exchange_kwargs:
-        exchange_kwargs["start_time"] = start_time
-    if "end_time" not in exchange_kwargs:
-        exchange_kwargs["end_time"] = end_time
-    trade_exchange = get_exchange(**exchange_kwargs)
+    if isinstance(exchange_kwargs, Exchange):
+        trade_exchange = exchange_kwargs
+    else:
+        exchange_kwargs = copy.copy(exchange_kwargs)
+        if "start_time" not in exchange_kwargs:
+            exchange_kwargs["start_time"] = start_time
+        if "end_time" not in exchange_kwargs:
+            exchange_kwargs["end_time"] = end_time
+        trade_exchange = get_exchange(**exchange_kwargs)
 
     common_infra = CommonInfrastructure(trade_account=trade_account, trade_exchange=trade_exchange)
     trade_strategy = init_instance_by_config(strategy, accept_types=BaseStrategy)
@@ -196,34 +217,36 @@ def get_strategy_executor(
 
 
 def backtest(
-    start_time,
-    end_time,
-    strategy,
-    executor,
-    benchmark="SH000300",
-    account=1e9,
-    exchange_kwargs={},
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    strategy: Union[str, dict, object, Path],
+    executor: Union[str, dict, object, Path],
+    benchmark: str = "SH000300",
+    account: Union[float, int, dict] = 1e9,
+    exchange_kwargs: dict = {},
     pos_type: str = "Position",
-):
-    """initialize the strategy and executor, then backtest function for the interaction of the outermost strategy and executor in the nested decision execution
+) -> Tuple[PORT_METRIC, INDICATOR_METRIC]:
+    """initialize the strategy and executor, then backtest function for the interaction of the outermost strategy and
+    executor in the nested decision execution
 
     Parameters
     ----------
-    start_time : pd.Timestamp|str
+    start_time : Union[pd.Timestamp, str]
         closed start time for backtest
         **NOTE**: This will be applied to the outmost executor's calendar.
-    end_time : pd.Timestamp|str
+    end_time : Union[pd.Timestamp, str]
         closed end time for backtest
         **NOTE**: This will be applied to the outmost executor's calendar.
         E.g. Executor[day](Executor[1min]),   setting `end_time == 20XX0301` will include all the minutes on 20XX0301
-    strategy : Union[str, dict, BaseStrategy]
-        for initializing outermost portfolio strategy. Please refer to the docs of init_instance_by_config for more information.
-    executor : Union[str, dict, BaseExecutor]
+    strategy : Union[str, dict, object, Path]
+        for initializing outermost portfolio strategy. Please refer to the docs of init_instance_by_config for more
+        information.
+    executor : Union[str, dict, object, Path]
         for initializing the outermost executor.
     benchmark: str
         the benchmark for reporting.
     account : Union[float, int, Position]
-        information for describing how to creating the account
+        information for describing how to create the account
         For `float` or `int`:
             Using Account with only initial cash
         For `Position`:
@@ -235,9 +258,9 @@ def backtest(
 
     Returns
     -------
-    portfolio_metrics_dict: Dict[PortfolioMetrics]
+    portfolio_dict: PORT_METRIC
         it records the trading portfolio_metrics information
-    indicator_dict: Dict[Indicator]
+    indicator_dict: INDICATOR_METRIC
         it computes the trading indicator
         It is organized in a dict format
 
@@ -252,21 +275,20 @@ def backtest(
         exchange_kwargs,
         pos_type=pos_type,
     )
-    portfolio_metrics, indicator = backtest_loop(start_time, end_time, trade_strategy, trade_executor)
-    return portfolio_metrics, indicator
+    return backtest_loop(start_time, end_time, trade_strategy, trade_executor)
 
 
 def collect_data(
-    start_time,
-    end_time,
-    strategy,
-    executor,
-    benchmark="SH000300",
-    account=1e9,
-    exchange_kwargs={},
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    strategy: Union[str, dict, object, Path],
+    executor: Union[str, dict, object, Path],
+    benchmark: str = "SH000300",
+    account: Union[float, int, dict] = 1e9,
+    exchange_kwargs: dict = {},
     pos_type: str = "Position",
-    return_value: dict = None,
-):
+    return_value: dict | None = None,
+) -> Generator[object, None, None]:
     """initialize the strategy and executor, then collect the trade decision data for rl training
 
     please refer to the docs of the backtest for the explanation of the parameters
@@ -291,7 +313,7 @@ def collect_data(
 
 def format_decisions(
     decisions: List[BaseTradeDecision],
-) -> Tuple[str, List[Tuple[BaseTradeDecision, Union[Tuple, None]]]]:
+) -> Optional[Tuple[str, List[Tuple[BaseTradeDecision, Union[Tuple, None]]]]]:
     """
     format the decisions collected by `qlib.backtest.collect_data`
     The decisions will be organized into a tree-like structure.
@@ -316,7 +338,7 @@ def format_decisions(
 
     cur_freq = decisions[0].strategy.trade_calendar.get_freq()
 
-    res = (cur_freq, [])
+    res: Tuple[str, list] = (cur_freq, [])
     last_dec_idx = 0
     for i, dec in enumerate(decisions[1:], 1):
         if dec.strategy.trade_calendar.get_freq() == cur_freq:
@@ -326,4 +348,4 @@ def format_decisions(
     return res
 
 
-__all__ = ["Order"]
+__all__ = ["Order", "backtest", "get_strategy_executor"]

qlib/backtest/account.py

@@ -1,15 +1,19 @@
 # Copyright (c) Microsoft Corporation.
 # Licensed under the MIT License.
 from __future__ import annotations
+
 import copy
-from typing import Dict, List, Tuple
-from qlib.utils import init_instance_by_config
+from typing import Dict, List, Optional, Tuple, cast
+
 import pandas as pd
 
-from .position import BasePosition
-from .report import PortfolioMetrics, Indicator
+from qlib.utils import init_instance_by_config
+
 from .decision import BaseTradeDecision, Order
 from .exchange import Exchange
+from .high_performance_ds import BaseOrderIndicator
+from .position import BasePosition
+from .report import Indicator, PortfolioMetrics
 
 """
 rtn & earning in the Account
@@ -34,40 +38,42 @@ class AccumulatedInfo:
     AccumulatedInfo should be shared across different levels
     """
 
-    def __init__(self):
+    def __init__(self) -> None:
         self.reset()
 
-    def reset(self):
-        self.rtn = 0  # accumulated return, do not consider cost
-        self.cost = 0  # accumulated cost
-        self.to = 0  # accumulated turnover
+    def reset(self) -> None:
+        self.rtn: float = 0.0  # accumulated return, do not consider cost
+        self.cost: float = 0.0  # accumulated cost
+        self.to: float = 0.0  # accumulated turnover
 
-    def add_return_value(self, value):
+    def add_return_value(self, value: float) -> None:
         self.rtn += value
 
-    def add_cost(self, value):
+    def add_cost(self, value: float) -> None:
         self.cost += value
 
-    def add_turnover(self, value):
+    def add_turnover(self, value: float) -> None:
         self.to += value
 
     @property
-    def get_return(self):
+    def get_return(self) -> float:
         return self.rtn
 
     @property
-    def get_cost(self):
+    def get_cost(self) -> float:
         return self.cost
 
     @property
-    def get_turnover(self):
+    def get_turnover(self) -> float:
         return self.to
 
 
 class Account:
     """
-    The correctness of the metrics of Account in nested execution depends on the shallow copy of `trade_account` in qlib/backtest/executor.py:NestedExecutor
-    Different level of executor has different Account object when calculating metrics. But the position object is shared cross all the Account object.
+    The correctness of the metrics of Account in nested execution depends on the shallow copy of `trade_account` in
+    qlib/backtest/executor.py:NestedExecutor
+    Different level of executor has different Account object when calculating metrics. But the position object is
+    shared cross all the Account object.
     """
 
     def __init__(
@@ -78,7 +84,7 @@ class Account:
         benchmark_config: dict = {},
         pos_type: str = "Position",
         port_metr_enabled: bool = True,
-    ):
+    ) -> None:
         """the trade account of backtest.
 
         Parameters
@@ -99,10 +105,10 @@ class Account:
 
         self._pos_type = pos_type
         self._port_metr_enabled = port_metr_enabled
-        self.benchmark_config = None  # avoid no attribute error
+        self.benchmark_config: dict = {}  # avoid no attribute error
         self.init_vars(init_cash, position_dict, freq, benchmark_config)
 
-    def init_vars(self, init_cash, position_dict, freq: str, benchmark_config: dict):
+    def init_vars(self, init_cash: float, position_dict: dict, freq: str, benchmark_config: dict) -> None:
         # 1) the following variables are shared by multiple layers
         # - you will see a shallow copy instead of deepcopy in the NestedExecutor;
         self.init_cash = init_cash
@@ -114,22 +120,22 @@ class Account:
                     "position_dict": position_dict,
                 },
                 "module_path": "qlib.backtest.position",
-            }
+            },
         )
         self.accum_info = AccumulatedInfo()
 
         # 2) following variables are not shared between layers
-        self.portfolio_metrics = None
-        self.hist_positions = {}
+        self.portfolio_metrics: Optional[PortfolioMetrics] = None
+        self.hist_positions: Dict[pd.Timestamp, BasePosition] = {}
         self.reset(freq=freq, benchmark_config=benchmark_config)
 
-    def is_port_metr_enabled(self):
+    def is_port_metr_enabled(self) -> bool:
         """
         Is portfolio-based metrics enabled.
         """
         return self._port_metr_enabled and not self.current_position.skip_update()
 
-    def reset_report(self, freq, benchmark_config):
+    def reset_report(self, freq: str, benchmark_config: dict) -> None:
         # portfolio related metrics
         if self.is_port_metr_enabled():
             # NOTE:
@@ -140,13 +146,15 @@ class Account:
             # fill stock value
             # The frequency of account may not align with the trading frequency.
             # This may result in obscure bugs when data quality is low.
-            if isinstance(self.benchmark_config, dict) and self.benchmark_config.get("start_time") is not None:
+            if isinstance(self.benchmark_config, dict) and "start_time" in self.benchmark_config:
                 self.current_position.fill_stock_value(self.benchmark_config["start_time"], self.freq)
 
         # trading related metrics(e.g. high-frequency trading)
         self.indicator = Indicator()
 
-    def reset(self, freq=None, benchmark_config=None, port_metr_enabled: bool = None):
+    def reset(
+        self, freq: str | None = None, benchmark_config: dict | None = None, port_metr_enabled: bool | None = None
+    ) -> None:
         """reset freq and report of account
 
         Parameters
@@ -155,6 +163,7 @@ class Account:
             frequency of account & report, by default None
         benchmark_config : {}, optional
             benchmark config of report, by default None
+        port_metr_enabled: bool
         """
         if freq is not None:
             self.freq = freq
@@ -165,13 +174,13 @@ class Account:
 
         self.reset_report(self.freq, self.benchmark_config)
 
-    def get_hist_positions(self):
+    def get_hist_positions(self) -> Dict[pd.Timestamp, BasePosition]:
         return self.hist_positions
 
-    def get_cash(self):
+    def get_cash(self) -> float:
         return self.current_position.get_cash()
 
-    def _update_state_from_order(self, order, trade_val, cost, trade_price):
+    def _update_state_from_order(self, order: Order, trade_val: float, cost: float, trade_price: float) -> None:
         if self.is_port_metr_enabled():
             # update turnover
             self.accum_info.add_turnover(trade_val)
@@ -191,13 +200,14 @@ class Account:
                 profit = self.current_position.get_stock_price(order.stock_id) * trade_amount - trade_val
                 self.accum_info.add_return_value(profit)  # note here do not consider cost
 
-    def update_order(self, order, trade_val, cost, trade_price):
+    def update_order(self, order: Order, trade_val: float, cost: float, trade_price: float) -> None:
         if self.current_position.skip_update():
             # TODO: supporting polymorphism for account
             # updating order for infinite position is meaningless
             return
 
-        # if stock is sold out, no stock price information in Position, then we should update account first, then update current position
+        # if stock is sold out, no stock price information in Position, then we should update account first,
+        # then update current position
         # if stock is bought, there is no stock in current position, update current, then update account
         # The cost will be subtracted from the cash at last. So the trading logic can ignore the cost calculation
         if order.direction == Order.SELL:
@@ -212,29 +222,40 @@ class Account:
             self.current_position.update_order(order, trade_val, cost, trade_price)
             self._update_state_from_order(order, trade_val, cost, trade_price)
 
-    def update_current_position(self, trade_start_time, trade_end_time, trade_exchange):
-        """update current to make rtn consistent with earning at the end of bar, and update holding bar count of stock"""
+    def update_current_position(
+        self,
+        trade_start_time: pd.Timestamp,
+        trade_end_time: pd.Timestamp,
+        trade_exchange: Exchange,
+    ) -> None:
+        """
+        Update current to make rtn consistent with earning at the end of bar, and update holding bar count of stock
+        """
         # update price for stock in the position and the profit from changed_price
         # NOTE: updating position does not only serve portfolio metrics, it also serve the strategy
+        assert self.current_position is not None
+
         if not self.current_position.skip_update():
             stock_list = self.current_position.get_stock_list()
             for code in stock_list:
-                # if suspend, no new price to be updated, profit is 0
+                # if suspended, no new price to be updated, profit is 0
                 if trade_exchange.check_stock_suspended(code, trade_start_time, trade_end_time):
                     continue
-                bar_close = trade_exchange.get_close(code, trade_start_time, trade_end_time)
+                bar_close = cast(float, trade_exchange.get_close(code, trade_start_time, trade_end_time))
                 self.current_position.update_stock_price(stock_id=code, price=bar_close)
             # update holding day count
             # NOTE: updating bar_count does not only serve portfolio metrics, it also serve the strategy
             self.current_position.add_count_all(bar=self.freq)
 
-    def update_portfolio_metrics(self, trade_start_time, trade_end_time):
+    def update_portfolio_metrics(self, trade_start_time: pd.Timestamp, trade_end_time: pd.Timestamp) -> None:
         """update portfolio_metrics"""
         # calculate earning
         # account_value - last_account_value
         # for the first trade date, account_value - init_cash
         # self.portfolio_metrics.is_empty() to judge is_first_trade_date
         # get last_account_value, last_total_cost, last_total_turnover
+        assert self.portfolio_metrics is not None
+
         if self.portfolio_metrics.is_empty():
             last_account_value = self.init_cash
             last_total_cost = 0
@@ -243,14 +264,16 @@ class Account:
             last_account_value = self.portfolio_metrics.get_latest_account_value()
             last_total_cost = self.portfolio_metrics.get_latest_total_cost()
             last_total_turnover = self.portfolio_metrics.get_latest_total_turnover()
+
         # get now_account_value, now_stock_value, now_earning, now_cost, now_turnover
         now_account_value = self.current_position.calculate_value()
         now_stock_value = self.current_position.calculate_stock_value()
         now_earning = now_account_value - last_account_value
         now_cost = self.accum_info.get_cost - last_total_cost
         now_turnover = self.accum_info.get_turnover - last_total_turnover
+
         # update portfolio_metrics for today
-        # judge whether the the trading is begin.
+        # judge whether the trading is begin.
         # and don't add init account state into portfolio_metrics, due to we don't have excess return in those days.
         self.portfolio_metrics.update_portfolio_metrics_record(
             trade_start_time=trade_start_time,
@@ -267,7 +290,7 @@ class Account:
             stock_value=now_stock_value,
         )
 
-    def update_hist_positions(self, trade_start_time):
+    def update_hist_positions(self, trade_start_time: pd.Timestamp) -> None:
         """update history position"""
         now_account_value = self.current_position.calculate_value()
         # set now_account_value to position
@@ -283,11 +306,11 @@ class Account:
         trade_exchange: Exchange,
         atomic: bool,
         outer_trade_decision: BaseTradeDecision,
-        trade_info: list = None,
-        inner_order_indicators: List[Dict[str, pd.Series]] = None,
-        decision_list: List[Tuple[BaseTradeDecision, pd.Timestamp, pd.Timestamp]] = None,
+        trade_info: list = [],
+        inner_order_indicators: List[BaseOrderIndicator] = [],
+        decision_list: List[Tuple[BaseTradeDecision, pd.Timestamp, pd.Timestamp]] = [],
         indicator_config: dict = {},
-    ):
+    ) -> None:
         """update trade indicators and order indicators in each bar end"""
         # TODO: will skip empty decisions make it faster?  `outer_trade_decision.empty():`
 
@@ -319,11 +342,11 @@ class Account:
         trade_exchange: Exchange,
         atomic: bool,
         outer_trade_decision: BaseTradeDecision,
-        trade_info: list = None,
-        inner_order_indicators: List[Dict[str, pd.Series]] = None,
-        decision_list: List[Tuple[BaseTradeDecision, pd.Timestamp, pd.Timestamp]] = None,
+        trade_info: list = [],
+        inner_order_indicators: List[BaseOrderIndicator] = [],
+        decision_list: List[Tuple[BaseTradeDecision, pd.Timestamp, pd.Timestamp]] = [],
         indicator_config: dict = {},
-    ):
+    ) -> None:
         """update account at each trading bar step
 
         Parameters
@@ -338,6 +361,8 @@ class Account:
             whether the trading executor is atomic, which means there is no higher-frequency trading executor inside it
             - if atomic is True, calculate the indicators with trade_info
             - else, aggregate indicators with inner indicators
+        outer_trade_decision: BaseTradeDecision
+            external trade decision
         trade_info : List[(Order, float, float, float)], optional
             trading information, by default None
             - necessary if atomic is True
@@ -377,9 +402,10 @@ class Account:
             indicator_config=indicator_config,
         )
 
-    def get_portfolio_metrics(self):
+    def get_portfolio_metrics(self) -> Tuple[pd.DataFrame, dict]:
         """get the history portfolio_metrics and positions instance"""
         if self.is_port_metr_enabled():
+            assert self.portfolio_metrics is not None
             _portfolio_metrics = self.portfolio_metrics.generate_portfolio_metrics_dataframe()
             _positions = self.get_hist_positions()
             return _portfolio_metrics, _positions

qlib/backtest/backtest.py

@@ -2,54 +2,81 @@
 # Licensed under the MIT License.
 
 from __future__ import annotations
+
+from typing import Dict, TYPE_CHECKING, Generator, Optional, Tuple, Union, cast
+
+import pandas as pd
+
 from qlib.backtest.decision import BaseTradeDecision
-from typing import TYPE_CHECKING
+from qlib.backtest.report import Indicator
 
 if TYPE_CHECKING:
     from qlib.strategy.base import BaseStrategy
     from qlib.backtest.executor import BaseExecutor
-from ..utils.time import Freq
+
 from tqdm.auto import tqdm
 
+from ..utils.time import Freq
 
-def backtest_loop(start_time, end_time, trade_strategy: BaseStrategy, trade_executor: BaseExecutor):
+
+PORT_METRIC = Dict[str, Tuple[pd.DataFrame, dict]]
+INDICATOR_METRIC = Dict[str, Tuple[pd.DataFrame, Indicator]]
+
+
+def backtest_loop(
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    trade_strategy: BaseStrategy,
+    trade_executor: BaseExecutor,
+) -> Tuple[PORT_METRIC, INDICATOR_METRIC]:
     """backtest function for the interaction of the outermost strategy and executor in the nested decision execution
 
     please refer to the docs of `collect_data_loop`
 
     Returns
     -------
-    portfolio_metrics: PortfolioMetrics
+    portfolio_dict: PORT_METRIC
         it records the trading portfolio_metrics information
-    indicator: Indicator
+    indicator_dict: INDICATOR_METRIC
         it computes the trading indicator
     """
-    return_value = {}
+    return_value: dict = {}
     for _decision in collect_data_loop(start_time, end_time, trade_strategy, trade_executor, return_value):
         pass
-    return return_value.get("portfolio_metrics"), return_value.get("indicator")
+
+    portfolio_dict = cast(PORT_METRIC, return_value.get("portfolio_dict"))
+    indicator_dict = cast(INDICATOR_METRIC, return_value.get("indicator_dict"))
+
+    return portfolio_dict, indicator_dict
 
 
 def collect_data_loop(
-    start_time, end_time, trade_strategy: BaseStrategy, trade_executor: BaseExecutor, return_value: dict = None
-):
+    start_time: Union[pd.Timestamp, str],
+    end_time: Union[pd.Timestamp, str],
+    trade_strategy: BaseStrategy,
+    trade_executor: BaseExecutor,
+    return_value: dict | None = None,
+    show_progress: bool = True,
+) -> Generator[BaseTradeDecision, Optional[BaseTradeDecision], None]:
     """Generator for collecting the trade decision data for rl training
 
     Parameters
     ----------
-    start_time : pd.Timestamp|str
+    start_time : Union[pd.Timestamp, str]
         closed start time for backtest
         **NOTE**: This will be applied to the outmost executor's calendar.
-    end_time : pd.Timestamp|str
+    end_time : Union[pd.Timestamp, str]
         closed end time for backtest
         **NOTE**: This will be applied to the outmost executor's calendar.
-        E.g. Executor[day](Executor[1min]),   setting `end_time == 20XX0301` will include all the minutes on 20XX0301
+        E.g. Executor[day](Executor[1min]), setting `end_time == 20XX0301` will include all the minutes on 20XX0301
     trade_strategy : BaseStrategy
         the outermost portfolio strategy
     trade_executor : BaseExecutor
         the outermost executor
     return_value : dict
         used for backtest_loop
+    show_progress: bool
+        whether to show execution progress
 
     Yields
     -------
@@ -59,23 +86,29 @@ def collect_data_loop(
     trade_executor.reset(start_time=start_time, end_time=end_time)
     trade_strategy.reset(level_infra=trade_executor.get_level_infra())
 
-    with tqdm(total=trade_executor.trade_calendar.get_trade_len(), desc="backtest loop") as bar:
+    disable = not show_progress
+    with tqdm(total=trade_executor.trade_calendar.get_trade_len(), desc="backtest loop", disable=disable) as bar:
         _execute_result = None
         while not trade_executor.finished():
             _trade_decision: BaseTradeDecision = trade_strategy.generate_trade_decision(_execute_result)
             _execute_result = yield from trade_executor.collect_data(_trade_decision, level=0)
+            trade_strategy.post_exe_step(_execute_result)
             bar.update(1)
+        trade_strategy.post_upper_level_exe_step()
 
     if return_value is not None:
         all_executors = trade_executor.get_all_executors()
-        all_portfolio_metrics = {
-            "{}{}".format(*Freq.parse(_executor.time_per_step)): _executor.trade_account.get_portfolio_metrics()
-            for _executor in all_executors
-            if _executor.trade_account.is_port_metr_enabled()
-        }
-        all_indicators = {}
-        for _executor in all_executors:
-            key = "{}{}".format(*Freq.parse(_executor.time_per_step))
-            all_indicators[key] = _executor.trade_account.get_trade_indicator().generate_trade_indicators_dataframe()
-            all_indicators[key + "_obj"] = _executor.trade_account.get_trade_indicator()
-        return_value.update({"portfolio_metrics": all_portfolio_metrics, "indicator": all_indicators})
+
+        portfolio_dict: PORT_METRIC = {}
+        indicator_dict: INDICATOR_METRIC = {}
+
+        for executor in all_executors:
+            key = "{}{}".format(*Freq.parse(executor.time_per_step))
+            if executor.trade_account.is_port_metr_enabled():
+                portfolio_dict[key] = executor.trade_account.get_portfolio_metrics()
+
+            indicator_df = executor.trade_account.get_trade_indicator().generate_trade_indicators_dataframe()
+            indicator_obj = executor.trade_account.get_trade_indicator()
+            indicator_dict[key] = (indicator_df, indicator_obj)
+
+        return_value.update({"portfolio_dict": portfolio_dict, "indicator_dict": indicator_dict})

qlib/backtest/decision.py

@@ -2,27 +2,33 @@
 # Licensed under the MIT License.
 
 from __future__ import annotations
-from enum import IntEnum
-from qlib.data.data import Cal
-from qlib.utils.time import concat_date_time, epsilon_change
-from qlib.log import get_module_logger
 
-from typing import ClassVar, Optional, Union, List, Tuple
+from abc import abstractmethod
+from datetime import time
+from enum import IntEnum
 
 # try to fix circular imports when enabling type hints
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, Any, ClassVar, Generic, List, Optional, Tuple, TypeVar, Union, cast
+
+from qlib.backtest.utils import TradeCalendarManager
+from qlib.data.data import Cal
+from qlib.log import get_module_logger
+from qlib.utils.time import concat_date_time, epsilon_change
 
 if TYPE_CHECKING:
     from qlib.strategy.base import BaseStrategy
     from qlib.backtest.exchange import Exchange
-from qlib.backtest.utils import TradeCalendarManager
+
+from dataclasses import dataclass
+
 import numpy as np
 import pandas as pd
-from dataclasses import dataclass
+
+DecisionType = TypeVar("DecisionType")
 
 
 class OrderDir(IntEnum):
-    # Order  direction
+    # Order direction
     SELL = 0
     BUY = 1
 
@@ -46,7 +52,7 @@ class Order:
     # - they are set by users and is time-invariant.
     stock_id: str
     amount: float  # `amount` is a non-negative and adjusted value
-    direction: int
+    direction: OrderDir
 
     # 2) time variant values:
     # - Users may want to set these values when using lower level APIs
@@ -61,8 +67,8 @@ class Order:
     # What the value should be about in all kinds of cases
     # - not tradable: the deal_amount == 0 , factor is None
     #    - the stock is suspended and the entire order fails. No cost for this order
-    # - dealed or partially dealed: deal_amount >= 0 and factor is not None
-    deal_amount: Optional[float] = None  # `deal_amount` is a non-negative value
+    # - dealt or partially dealt: deal_amount >= 0 and factor is not None
+    deal_amount: float = 0.0  # `deal_amount` is a non-negative value
     factor: Optional[float] = None
 
     # TODO:
@@ -74,10 +80,10 @@ class Order:
     SELL: ClassVar[OrderDir] = OrderDir.SELL
     BUY: ClassVar[OrderDir] = OrderDir.BUY
 
-    def __post_init__(self):
+    def __post_init__(self) -> None:
         if self.direction not in {Order.SELL, Order.BUY}:
             raise NotImplementedError("direction not supported, `Order.SELL` for sell, `Order.BUY` for buy")
-        self.deal_amount = 0
+        self.deal_amount = 0.0
         self.factor = None
 
     @property
@@ -99,7 +105,7 @@ class Order:
         return self.deal_amount * self.sign
 
     @property
-    def sign(self) -> float:
+    def sign(self) -> int:
         """
         return the sign of trading
         - `+1` indicates buying
@@ -112,15 +118,12 @@ class Order:
         if isinstance(direction, OrderDir):
             return direction
         elif isinstance(direction, (int, float, np.integer, np.floating)):
-            if direction > 0:
-                return Order.BUY
-            else:
-                return Order.SELL
+            return Order.BUY if direction > 0 else Order.SELL
         elif isinstance(direction, str):
-            dl = direction.lower()
-            if dl.strip() == "sell":
+            dl = direction.lower().strip()
+            if dl == "sell":
                 return OrderDir.SELL
-            elif dl.strip() == "buy":
+            elif dl == "buy":
                 return OrderDir.BUY
             else:
                 raise NotImplementedError(f"This type of input is not supported")
@@ -132,20 +135,35 @@ class Order:
         else:
             raise NotImplementedError(f"This type of input is not supported")
 
+    @property
+    def key_by_day(self) -> tuple:
+        """A hashable & unique key to identify this order, under the granularity in day."""
+        return self.stock_id, self.date, self.direction
+
+    @property
+    def key(self) -> tuple:
+        """A hashable & unique key to identify this order."""
+        return self.stock_id, self.start_time, self.end_time, self.direction
+
+    @property
+    def date(self) -> pd.Timestamp:
+        """Date of the order."""
+        return pd.Timestamp(self.start_time.replace(hour=0, minute=0, second=0))
+
 
 class OrderHelper:
     """
     Motivation
     - Make generating order easier
         - User may have no knowledge about the adjust-factor information about the system.
-        - It involves to much interaction with the exchange when generating orders.
+        - It involves too much interaction with the exchange when generating orders.
     """
 
-    def __init__(self, exchange: Exchange):
+    def __init__(self, exchange: Exchange) -> None:
         self.exchange = exchange
 
+    @staticmethod
     def create(
-        self,
         code: str,
         amount: float,
         direction: OrderDir,
@@ -175,21 +193,18 @@ class OrderHelper:
         Order:
             The created order
         """
-        if start_time is not None:
-            start_time = pd.Timestamp(start_time)
-        if end_time is not None:
-            end_time = pd.Timestamp(end_time)
         # NOTE: factor is a value belongs to the results section. User don't have to care about it when creating orders
         return Order(
             stock_id=code,
             amount=amount,
-            start_time=start_time,
-            end_time=end_time,
+            start_time=None if start_time is None else pd.Timestamp(start_time),
+            end_time=None if end_time is None else pd.Timestamp(end_time),
             direction=direction,
         )
 
 
 class TradeRange:
+    @abstractmethod
     def __call__(self, trade_calendar: TradeCalendarManager) -> Tuple[int, int]:
         """
         This method will be call with following way
@@ -216,6 +231,7 @@ class TradeRange:
         """
         raise NotImplementedError(f"Please implement the `__call__` method")
 
+    @abstractmethod
     def clip_time_range(self, start_time: pd.Timestamp, end_time: pd.Timestamp) -> Tuple[pd.Timestamp, pd.Timestamp]:
         """
         Parameters
@@ -234,56 +250,58 @@ class TradeRange:
 
 
 class IdxTradeRange(TradeRange):
-    def __init__(self, start_idx: int, end_idx: int):
+    def __init__(self, start_idx: int, end_idx: int) -> None:
         self._start_idx = start_idx
         self._end_idx = end_idx
 
-    def __call__(self, trade_calendar: TradeCalendarManager = None) -> Tuple[int, int]:
+    def __call__(self, trade_calendar: TradeCalendarManager | None = None) -> Tuple[int, int]:
         return self._start_idx, self._end_idx
 
+    def clip_time_range(self, start_time: pd.Timestamp, end_time: pd.Timestamp) -> Tuple[pd.Timestamp, pd.Timestamp]:
+        raise NotImplementedError
+
 
 class TradeRangeByTime(TradeRange):
     """This is a helper function for make decisions"""
 
-    def __init__(self, start_time: str, end_time: str):
+    def __init__(self, start_time: str | time, end_time: str | time) -> None:
         """
         This is a callable class.
 
         **NOTE**:
-        - It is designed for minute-bar for intraday trading!!!!!
+        - It is designed for minute-bar for intra-day trading!!!!!
         - Both start_time and end_time are **closed** in the range
 
         Parameters
         ----------
-        start_time : str
+        start_time : str | time
             e.g. "9:30"
-        end_time : str
+        end_time : str | time
             e.g. "14:30"
         """
-        self.start_time = pd.Timestamp(start_time).time()
-        self.end_time = pd.Timestamp(end_time).time()
+        self.start_time = pd.Timestamp(start_time).time() if isinstance(start_time, str) else start_time
+        self.end_time = pd.Timestamp(end_time).time() if isinstance(end_time, str) else end_time
         assert self.start_time < self.end_time
 
-    def __call__(self, trade_calendar: TradeCalendarManager = None) -> Tuple[int, int]:
+    def __call__(self, trade_calendar: TradeCalendarManager) -> Tuple[int, int]:
         if trade_calendar is None:
             raise NotImplementedError("trade_calendar is necessary for getting TradeRangeByTime.")
-        start = trade_calendar.start_time
-        val_start, val_end = concat_date_time(start.date(), self.start_time), concat_date_time(
-            start.date(), self.end_time
-        )
+
+        start_date = trade_calendar.start_time.date()
+        val_start, val_end = concat_date_time(start_date, self.start_time), concat_date_time(start_date, self.end_time)
         return trade_calendar.get_range_idx(val_start, val_end)
 
     def clip_time_range(self, start_time: pd.Timestamp, end_time: pd.Timestamp) -> Tuple[pd.Timestamp, pd.Timestamp]:
         start_date = start_time.date()
         val_start, val_end = concat_date_time(start_date, self.start_time), concat_date_time(start_date, self.end_time)
         # NOTE: `end_date` should not be used. Because the `end_date` is for slicing. It may be in the next day
-        # Assumption: start_time and end_time is for intraday trading. So it is OK for only using start_date
+        # Assumption: start_time and end_time is for intra-day trading. So it is OK for only using start_date
         return max(val_start, start_time), min(val_end, end_time)
 
 
-class BaseTradeDecision:
+class BaseTradeDecision(Generic[DecisionType]):
     """
-    Trade decisions ara made by strategy and executed by exeuter
+    Trade decisions are made by strategy and executed by executor
 
     Motivation:
         Here are several typical scenarios for `BaseTradeDecision`
@@ -297,7 +315,7 @@ class BaseTradeDecision:
         2. Same as `case 1.3`
     """
 
-    def __init__(self, strategy: BaseStrategy, trade_range: Union[Tuple[int, int], TradeRange] = None):
+    def __init__(self, strategy: BaseStrategy, trade_range: Union[Tuple[int, int], TradeRange, None] = None) -> None:
         """
         Parameters
         ----------
@@ -316,20 +334,21 @@ class BaseTradeDecision:
         """
         self.strategy = strategy
         self.start_time, self.end_time = strategy.trade_calendar.get_step_time()
-        self.total_step = None  # upper strategy has no knowledge about the sub executor before `_init_sub_trading`
-        if isinstance(trade_range, Tuple):
+        # upper strategy has no knowledge about the sub executor before `_init_sub_trading`
+        self.total_step: Optional[int] = None
+        if isinstance(trade_range, tuple):
             # for Tuple[int, int]
             trade_range = IdxTradeRange(*trade_range)
-        self.trade_range: TradeRange = trade_range
+        self.trade_range: Optional[TradeRange] = trade_range
 
-    def get_decision(self) -> List[object]:
+    def get_decision(self) -> List[DecisionType]:
         """
         get the **concrete decision**  (e.g. execution orders)
         This will be called by the inner strategy
 
         Returns
         -------
-        List[object]:
+        List[DecisionType:
             The decision result. Typically it is some orders
             Example:
                 []:
@@ -339,7 +358,7 @@ class BaseTradeDecision:
         """
         raise NotImplementedError(f"This type of input is not supported")
 
-    def update(self, trade_calendar: TradeCalendarManager) -> Union["BaseTradeDecision", None]:
+    def update(self, trade_calendar: TradeCalendarManager) -> Optional[BaseTradeDecision]:
         """
         Be called at the **start** of each step.
 
@@ -354,10 +373,8 @@ class BaseTradeDecision:
 
         Returns
         -------
-        None:
-            No update, use previous decision(or unavailable)
         BaseTradeDecision:
-            New update, use new decision
+            New update, use new decision. If no updates, return None (use previous decision (or unavailable))
         """
         # purpose 1)
         self.total_step = trade_calendar.get_trade_len()
@@ -365,13 +382,13 @@ class BaseTradeDecision:
         # purpose 2)
         return self.strategy.update_trade_decision(self, trade_calendar)
 
-    def _get_range_limit(self, **kwargs) -> Tuple[int, int]:
+    def _get_range_limit(self, **kwargs: Any) -> Tuple[int, int]:
         if self.trade_range is not None:
-            return self.trade_range(trade_calendar=kwargs.get("inner_calendar"))
+            return self.trade_range(trade_calendar=cast(TradeCalendarManager, kwargs.get("inner_calendar")))
         else:
             raise NotImplementedError("The decision didn't provide an index range")
 
-    def get_range_limit(self, **kwargs) -> Tuple[int, int]:
+    def get_range_limit(self, **kwargs: Any) -> Tuple[int, int]:
         """
         return the expected step range for limiting the decision execution time
         Both left and right are **closed**
@@ -412,21 +429,22 @@ class BaseTradeDecision:
         """
         try:
             _start_idx, _end_idx = self._get_range_limit(**kwargs)
-        except NotImplementedError:
+        except NotImplementedError as e:
             if "default_value" in kwargs:
                 return kwargs["default_value"]
             else:
                 # Default to get full index
-                raise NotImplementedError(f"The decision didn't provide an index range") from NotImplementedError
+                raise NotImplementedError(f"The decision didn't provide an index range") from e
 
         # clip index
         if getattr(self, "total_step", None) is not None:
             # if `self.update` is called.
             # Then the _start_idx, _end_idx should be clipped
+            assert self.total_step is not None
             if _start_idx < 0 or _end_idx >= self.total_step:
                 logger = get_module_logger("decision")
                 logger.warning(
-                    f"[{_start_idx},{_end_idx}] go beyoud the total_step({self.total_step}), it will be clipped"
+                    f"[{_start_idx},{_end_idx}] go beyond the total_step({self.total_step}), it will be clipped.",
                 )
                 _start_idx, _end_idx = max(0, _start_idx), min(self.total_step - 1, _end_idx)
         return _start_idx, _end_idx
@@ -444,7 +462,7 @@ class BaseTradeDecision:
         Parameters
         ----------
         rtype: str
-            - "full": return the full limitation of the deicsion in the day
+            - "full": return the full limitation of the decision in the day
             - "step": return the limitation of current step
 
         raise_error: bool
@@ -497,11 +515,10 @@ class BaseTradeDecision:
                 return True
         return True
 
-    def mod_inner_decision(self, inner_trade_decision: BaseTradeDecision):
+    def mod_inner_decision(self, inner_trade_decision: BaseTradeDecision) -> None:
         """
-
         This method will be called on the inner_trade_decision after it is generated.
-        `inner_trade_decision` will be changed **inplaced**.
+        `inner_trade_decision` will be changed **inplace**.
 
         Motivation of the `mod_inner_decision`
         - Leave a hook for outer decision to affect the decision generated by the inner strategy
@@ -519,29 +536,61 @@ class BaseTradeDecision:
             inner_trade_decision.trade_range = self.trade_range
 
 
-class EmptyTradeDecision(BaseTradeDecision):
+class EmptyTradeDecision(BaseTradeDecision[object]):
+    def get_decision(self) -> List[object]:
+        return []
+
     def empty(self) -> bool:
         return True
 
 
-class TradeDecisionWO(BaseTradeDecision):
+class TradeDecisionWO(BaseTradeDecision[Order]):
     """
     Trade Decision (W)ith (O)rder.
     Besides, the time_range is also included.
     """
 
-    def __init__(self, order_list: List[Order], strategy: BaseStrategy, trade_range: Tuple[int, int] = None):
+    def __init__(
+        self,
+        order_list: List[Order],
+        strategy: BaseStrategy,
+        trade_range: Union[Tuple[int, int], TradeRange, None] = None,
+    ) -> None:
         super().__init__(strategy, trade_range=trade_range)
-        self.order_list = order_list
+        self.order_list = cast(List[Order], order_list)
         start, end = strategy.trade_calendar.get_step_time()
         for o in order_list:
+            assert isinstance(o, Order)
             if o.start_time is None:
                 o.start_time = start
             if o.end_time is None:
                 o.end_time = end
 
-    def get_decision(self) -> List[object]:
+    def get_decision(self) -> List[Order]:
         return self.order_list
 
     def __repr__(self) -> str:
-        return f"class: {self.__class__.__name__}; strategy: {self.strategy}; trade_range: {self.trade_range}; order_list[{len(self.order_list)}]"
+        return (
+            f"class: {self.__class__.__name__}; "
+            f"strategy: {self.strategy}; "
+            f"trade_range: {self.trade_range}; "
+            f"order_list[{len(self.order_list)}]"
+        )
+
+
+class TradeDecisionWithDetails(TradeDecisionWO):
+    """
+    Decision with detail information.
+    Detail information is used to generate execution reports.
+    """
+
+    def __init__(
+        self,
+        order_list: List[Order],
+        strategy: BaseStrategy,
+        trade_range: Optional[Tuple[int, int]] = None,
+        details: Optional[Any] = None,
+    ) -> None:
+        super().__init__(order_list, strategy, trade_range)
+
+        self.details = details

qlib/backtest/exchange.py

@@ -1,45 +1,58 @@
 # Copyright (c) Microsoft Corporation.
 # Licensed under the MIT License.
 from __future__ import annotations
+
 from collections import defaultdict
-from typing import TYPE_CHECKING
-from typing import List, Tuple, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Type, Union, cast
+
+from ..utils.index_data import IndexData
 
 if TYPE_CHECKING:
     from .account import Account
 
-from qlib.backtest.position import BasePosition, Position
 import random
+
 import numpy as np
 import pandas as pd
 
-from ..data.data import D
+from qlib.backtest.position import BasePosition
+
 from ..config import C
-from ..constant import REG_CN
+from ..constant import REG_CN, REG_TW
+from ..data.data import D
 from ..log import get_module_logger
 from .decision import Order, OrderDir, OrderHelper
 from .high_performance_ds import BaseQuote, NumpyQuote
 
 
 class Exchange:
+    # `quote_df` is a pd.DataFrame class that contains basic information for backtesting
+    # After some processing, the data will later be maintained by `quote_cls` object for faster data retrieving.
+    # Some conventions for `quote_df`
+    # - $close is for calculating the total value at end of each day.
+    #   - if $close is None, the stock on that day is regarded as suspended.
+    # - $factor is for rounding to the trading unit;
+    #   - if any $factor is missing when $close exists, trading unit rounding will be disabled
+    quote_df: pd.DataFrame
+
     def __init__(
         self,
-        freq="day",
-        start_time=None,
-        end_time=None,
-        codes="all",
-        deal_price: Union[str, Tuple[str], List[str]] = None,
-        subscribe_fields=[],
+        freq: str = "day",
+        start_time: Union[pd.Timestamp, str] = None,
+        end_time: Union[pd.Timestamp, str] = None,
+        codes: Union[list, str] = "all",
+        deal_price: Union[str, Tuple[str, str], List[str], None] = None,
+        subscribe_fields: list = [],
         limit_threshold: Union[Tuple[str, str], float, None] = None,
-        volume_threshold=None,
-        open_cost=0.0015,
-        close_cost=0.0025,
-        min_cost=5,
-        impact_cost=0.0,
-        extra_quote=None,
-        quote_cls=NumpyQuote,
-        **kwargs,
-    ):
+        volume_threshold: Union[tuple, dict, None] = None,
+        open_cost: float = 0.0015,
+        close_cost: float = 0.0025,
+        min_cost: float = 5.0,
+        impact_cost: float = 0.0,
+        extra_quote: pd.DataFrame = None,
+        quote_cls: Type[BaseQuote] = NumpyQuote,
+        **kwargs: Any,
+    ) -> None:
         """__init__
         :param freq:             frequency of data
         :param start_time:       closed start time for backtest
@@ -72,11 +85,12 @@ class Exchange:
                                  ]
                                 1) ("cum" or "current", limit_str) denotes a single volume limit.
                                     - limit_str is qlib data expression which is allowed to define your own Operator.
-                                    Please refer to qlib/contrib/ops/high_freq.py, here are any custom operator for high frequency,
-                                    such as DayCumsum. !!!NOTE: if you want you use the custom operator, you need to
-                                    register it in qlib_init.
-                                    - "cum" means that this is a cumulative value over time, such as cumulative market volume.
-                                    So when it is used as a volume limit, it is necessary to subtract the dealt amount.
+                                    Please refer to qlib/contrib/ops/high_freq.py, here are any custom operator for
+                                    high frequency, such as DayCumsum. !!!NOTE: if you want you use the custom
+                                    operator, you need to register it in qlib_init.
+                                    - "cum" means that this is a cumulative value over time, such as cumulative market
+                                    volume. So when it is used as a volume limit, it is necessary to subtract the dealt
+                                    amount.
                                     - "current" means that this is a real-time value and will not accumulate over time,
                                     so it can be directly used as a capacity limit.
                                     e.g. ("cum", "0.2 * DayCumsum($volume, '9:45', '14:45')"), ("current", "$bidV1")
@@ -84,7 +98,7 @@ class Exchange:
                                 "buy" means the volume limits of buying. "sell" means the volume limits of selling.
                                 Different volume limits will be aggregated with min(). If volume_threshold is only
                                 ("cum" or "current", limit_str) instead of a dict, the volume limits are for
-                                both by deault. In other words, it is same as {"all": ("cum" or "current", limit_str)}.
+                                both by default. In other words, it is same as {"all": ("cum" or "current", limit_str)}.
                                 3) e.g. "volume_threshold": {
                                             "all": ("cum", "0.2 * DayCumsum($volume, '9:45', '14:45')"),
                                             "buy": ("current", "$askV1"),
@@ -104,13 +118,14 @@ class Exchange:
                                             Necessary fields:
                                                 $close is for calculating the total value at end of each day.
                                             Optional fields:
-                                                $volume is only necessary when we limit the trade amount or calculate PA(vwap) indicator
+                                                $volume is only necessary when we limit the trade amount or calculate
+                                                PA(vwap) indicator
                                                 $vwap is only necessary when we use the $vwap price as the deal price
                                                 $factor is for rounding to the trading unit
-                                                limit_sell will be set to False by default(False indicates we can sell this
-                                                target on this day).
-                                                limit_buy will be set to False by default(False indicates we can buy this
-                                                target on this day).
+                                                limit_sell will be set to False by default (False indicates we can sell
+                                                this target on this day).
+                                                limit_buy will be set to False by default (False indicates we can buy
+                                                this target on this day).
                                     index: MultipleIndex(instrument, pd.Datetime)
         """
         self.freq = freq
@@ -126,17 +141,17 @@ class Exchange:
         if deal_price is None:
             deal_price = C.deal_price
 
-        # we have some verbose information here. So logging is enable
+        # we have some verbose information here. So logging is enabled
         self.logger = get_module_logger("online operator")
 
         # TODO: the quote, trade_dates, codes are not necessary.
         # It is just for performance consideration.
         self.limit_type = self._get_limit_type(limit_threshold)
         if limit_threshold is None:
-            if C.region == REG_CN:
+            if C.region in [REG_CN, REG_TW]:
                 self.logger.warning(f"limit_threshold not set. The stocks hit the limit may be bought/sold")
-        elif self.limit_type == self.LT_FLT and abs(limit_threshold) > 0.1:
-            if C.region == REG_CN:
+        elif self.limit_type == self.LT_FLT and abs(cast(float, limit_threshold)) > 0.1:
+            if C.region in [REG_CN, REG_TW]:
                 self.logger.warning(f"limit_threshold may not be set to a reasonable value")
 
         if isinstance(deal_price, str):
@@ -144,7 +159,7 @@ class Exchange:
                 deal_price = "$" + deal_price
             self.buy_price = self.sell_price = deal_price
         elif isinstance(deal_price, (tuple, list)):
-            self.buy_price, self.sell_price = deal_price
+            self.buy_price, self.sell_price = cast(Tuple[str, str], deal_price)
         else:
             raise NotImplementedError(f"This type of input is not supported")
 
@@ -153,6 +168,7 @@ class Exchange:
         self.codes = codes
         # Necessary fields
         # $close is for calculating the total value at end of each day.
+        # - if $close is None, the stock on that day is regarded as suspended.
         # $factor is for rounding to the trading unit
         # $change is for calculating the limit of the stock
 
@@ -161,10 +177,10 @@ class Exchange:
 
         necessary_fields = {self.buy_price, self.sell_price, "$close", "$change", "$factor", "$volume"}
         if self.limit_type == self.LT_TP_EXP:
+            assert isinstance(limit_threshold, tuple) or (isinstance(limit_threshold, list) and len(limit_threshold) == 2)
             for exp in limit_threshold:
                 necessary_fields.add(exp)
-        all_fields = necessary_fields | vol_lt_fields
-        all_fields = list(all_fields | set(subscribe_fields))
+        all_fields = list(necessary_fields | set(vol_lt_fields) | set(subscribe_fields))
 
         self.all_fields = all_fields
 
@@ -182,23 +198,28 @@ class Exchange:
         self.quote_cls = quote_cls
         self.quote: BaseQuote = self.quote_cls(self.quote_df, freq)
 
-    def get_quote_from_qlib(self):
+    def get_quote_from_qlib(self) -> None:
         # get stock data from qlib
         if len(self.codes) == 0:
             self.codes = D.instruments()
         self.quote_df = D.features(
-            self.codes, self.all_fields, self.start_time, self.end_time, freq=self.freq, disk_cache=True
-        ).dropna(subset=["$close"])
+            self.codes,
+            self.all_fields,
+            self.start_time,
+            self.end_time,
+            freq=self.freq,
+            disk_cache=True,
+        )
         self.quote_df.columns = self.all_fields
 
         # check buy_price data and sell_price data
-        for attr in "buy_price", "sell_price":
+        for attr in ("buy_price", "sell_price"):
             pstr = getattr(self, attr)  # price string
             if self.quote_df[pstr].isna().any():
                 self.logger.warning("{} field data contains nan.".format(pstr))
 
         # update trade_w_adj_price
-        if self.quote_df["$factor"].isna().any():
+        if (self.quote_df["$factor"].isna() & ~self.quote_df["$close"].isna()).any():
             # The 'factor.day.bin' file not exists, and `factor` field contains `nan`
             # Use adjusted price
             self.trade_w_adj_price = True
@@ -234,13 +255,16 @@ class Exchange:
             assert set(self.extra_quote.columns) == set(self.quote_df.columns) - {"$change"}
             self.quote_df = pd.concat([self.quote_df, self.extra_quote], sort=False, axis=0)
 
-    LT_TP_EXP = "(exp)"  # Tuple[str, str]
-    LT_FLT = "float"  # float
-    LT_NONE = "none"  # none
+    LT_TP_EXP = "(exp)"  # Tuple[str, str]:  the limitation is calculated by a Qlib expression.
+    LT_FLT = "float"  # float:  the trading limitation is based on `abs($change) < limit_threshold`
+    LT_NONE = "none"  # none:  there is no trading limitation
 
-    def _get_limit_type(self, limit_threshold):
+    def _get_limit_type(self, limit_threshold: Union[tuple, float, None]) -> str:
         """get limit type"""
-        if isinstance(limit_threshold, Tuple):
+        if isinstance(limit_threshold, tuple):
+            return self.LT_TP_EXP
+        if isinstance(limit_threshold, list):
+            assert len(limit_threshold) == 2
             return self.LT_TP_EXP
         elif isinstance(limit_threshold, float):
             return self.LT_FLT
@@ -249,23 +273,31 @@ class Exchange:
         else:
             raise NotImplementedError(f"This type of `limit_threshold` is not supported")
 
-    def _update_limit(self, limit_threshold):
+    def _update_limit(self, limit_threshold: Union[Tuple, float, None]) -> None:
+        # $close may contain NaN, the nan indicates that the stock is not tradable at that timestamp
+        suspended = self.quote_df["$close"].isna()
         # check limit_threshold
         limit_type = self._get_limit_type(limit_threshold)
         if limit_type == self.LT_NONE:
-            self.quote_df["limit_buy"] = False
-            self.quote_df["limit_sell"] = False
+            self.quote_df["limit_buy"] = suspended
+            self.quote_df["limit_sell"] = suspended
         elif limit_type == self.LT_TP_EXP:
             # set limit
-            self.quote_df["limit_buy"] = self.quote_df[limit_threshold[0]]
-            self.quote_df["limit_sell"] = self.quote_df[limit_threshold[1]]
+            limit_threshold = cast(tuple, limit_threshold)
+            # astype bool is necessary, because quote_df is an expression and could be float
+            self.quote_df["limit_buy"] = self.quote_df[limit_threshold[0]].astype("bool") | suspended
+            self.quote_df["limit_sell"] = self.quote_df[limit_threshold[1]].astype("bool") | suspended
         elif limit_type == self.LT_FLT:
-            self.quote_df["limit_buy"] = self.quote_df["$change"].ge(limit_threshold)
-            self.quote_df["limit_sell"] = self.quote_df["$change"].le(-limit_threshold)  # pylint: disable=E1130
+            limit_threshold = cast(float, limit_threshold)
+            self.quote_df["limit_buy"] = self.quote_df["$change"].ge(limit_threshold) | suspended
+            self.quote_df["limit_sell"] = (
+                self.quote_df["$change"].le(-limit_threshold) | suspended
+            )  # pylint: disable=E1130
 
-    def _get_vol_limit(self, volume_threshold):
+    @staticmethod
+    def _get_vol_limit(volume_threshold: Union[tuple, dict, None]) -> Tuple[Optional[list], Optional[list], set]:
         """
-        preproccess the volume limit.
+        preprocess the volume limit.
         get the fields need to get from qlib.
         get the volume limit list of buying and selling which is composed of all limits.
         Parameters
@@ -295,9 +327,8 @@ class Exchange:
             volume_threshold = {"all": volume_threshold}
 
         assert isinstance(volume_threshold, dict)
-        for key in volume_threshold:
-            vol_limit = volume_threshold[key]
-            assert isinstance(vol_limit, tuple)
+        for key, vol_limit in volume_threshold.items():
+            assert isinstance(vol_limit, tuple) or (isinstance(vol_limit, list) and len(vol_limit) == 2)
             fields.add(vol_limit[1])
 
             if key in ("buy", "all"):
@@ -307,60 +338,96 @@ class Exchange:
 
         return buy_vol_limit, sell_vol_limit, fields
 
-    def check_stock_limit(self, stock_id, start_time, end_time, direction=None):
+    def check_stock_limit(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        direction: int | None = None,
+    ) -> bool:
         """
         Parameters
         ----------
+        stock_id : str
+        start_time: pd.Timestamp
+        end_time: pd.Timestamp
         direction : int, optional
             trade direction, by default None
             - if direction is None, check if tradable for buying and selling.
             - if direction == Order.BUY, check the if tradable for buying
             - if direction == Order.SELL, check the sell limit for selling.
+
+        Returns
+        -------
+        True: the trading of the stock is limited (maybe hit the highest/lowest price), hence the stock is not tradable
+        False: the trading of the stock is not limited, hence the stock may be tradable
         """
+        # NOTE:
+        # **all** is used when checking limitation.
+        # For example, the stock trading is limited in a day if every minute is limited in a day if every minute is limited.
         if direction is None:
+            # The trading limitation is related to the trading direction
+            # if the direction is not provided, then any limitation from buy or sell will result in trading limitation
             buy_limit = self.quote.get_data(stock_id, start_time, end_time, field="limit_buy", method="all")
             sell_limit = self.quote.get_data(stock_id, start_time, end_time, field="limit_sell", method="all")
-            return buy_limit or sell_limit
+            return bool(buy_limit or sell_limit)
         elif direction == Order.BUY:
-            return self.quote.get_data(stock_id, start_time, end_time, field="limit_buy", method="all")
+            return cast(bool, self.quote.get_data(stock_id, start_time, end_time, field="limit_buy", method="all"))
         elif direction == Order.SELL:
-            return self.quote.get_data(stock_id, start_time, end_time, field="limit_sell", method="all")
+            return cast(bool, self.quote.get_data(stock_id, start_time, end_time, field="limit_sell", method="all"))
         else:
             raise ValueError(f"direction {direction} is not supported!")
 
-    def check_stock_suspended(self, stock_id, start_time, end_time):
+    def check_stock_suspended(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+    ) -> bool:
+        """if stock is suspended(hence not tradable), True will be returned"""
         # is suspended
         if stock_id in self.quote.get_all_stock():
-            return self.quote.get_data(stock_id, start_time, end_time, "$close") is None
+            # suspended stocks are represented by None $close stock
+            # The $close may contain NaN,
+            close = self.quote.get_data(stock_id, start_time, end_time, "$close")
+            if close is None:
+                # if no close record exists
+                return True
+            elif isinstance(close, IndexData):
+                # **any** non-NaN $close represents trading opportunity may exist
+                #  if all returned is nan, then the stock is suspended
+                return cast(bool, cast(IndexData, close).isna().all())
+            else:
+                # it is single value, make sure is not None
+                return np.isnan(close)
         else:
+            # if the stock is not in the stock list, then it is not tradable and regarded as suspended
             return True
 
-    def is_stock_tradable(self, stock_id, start_time, end_time, direction=None):
+    def is_stock_tradable(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        direction: int | None = None,
+    ) -> bool:
         # check if stock can be traded
-        # same as check in check_order
-        if self.check_stock_suspended(stock_id, start_time, end_time) or self.check_stock_limit(
-            stock_id, start_time, end_time, direction
-        ):
-            return False
-        else:
-            return True
+        return not (
+            self.check_stock_suspended(stock_id, start_time, end_time)
+            or self.check_stock_limit(stock_id, start_time, end_time, direction)
+        )
 
-    def check_order(self, order):
+    def check_order(self, order: Order) -> bool:
         # check limit and suspended
-        if self.check_stock_suspended(order.stock_id, order.start_time, order.end_time) or self.check_stock_limit(
-            order.stock_id, order.start_time, order.end_time, order.direction
-        ):
-            return False
-        else:
-            return True
+        return self.is_stock_tradable(order.stock_id, order.start_time, order.end_time, order.direction)
 
     def deal_order(
         self,
-        order,
-        trade_account: Account = None,
-        position: BasePosition = None,
-        dealt_order_amount: defaultdict = defaultdict(float),
-    ):
+        order: Order,
+        trade_account: Account | None = None,
+        position: BasePosition | None = None,
+        dealt_order_amount: Dict[str, float] = defaultdict(float),
+    ) -> Tuple[float, float, float]:
         """
         Deal order when the actual transaction
         the results section in `Order` will be changed.
@@ -371,9 +438,9 @@ class Exchange:
         :return: trade_val, trade_cost, trade_price
         """
         # check order first.
-        if self.check_order(order) is False:
+        if not self.check_order(order):
             order.deal_amount = 0.0
-            # using np.nan instead of None to make it more convenient to should the value in format string
+            # using np.nan instead of None to make it more convenient to show the value in format string
             self.logger.debug(f"Order failed due to trading limitation: {order}")
             return 0.0, 0.0, np.nan
 
@@ -382,7 +449,9 @@ class Exchange:
 
         # NOTE: order will be changed in this function
         trade_price, trade_val, trade_cost = self._calc_trade_info_by_order(
-            order, trade_account.current_position if trade_account else position, dealt_order_amount
+            order,
+            trade_account.current_position if trade_account else position,
+            dealt_order_amount,
         )
         if trade_val > 1e-5:
             # If the order can only be deal 0 value. Nothing to be updated
@@ -396,23 +465,50 @@ class Exchange:
 
         return trade_val, trade_cost, trade_price
 
-    def get_quote_info(self, stock_id, start_time, end_time, method="ts_data_last"):
-        return self.quote.get_data(stock_id, start_time, end_time, method=method)
+    def get_quote_info(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        field: str,
+        method: str = "ts_data_last",
+    ) -> Union[None, int, float, bool, IndexData]:
+        return self.quote.get_data(stock_id, start_time, end_time, field=field, method=method)
 
-    def get_close(self, stock_id, start_time, end_time, method="ts_data_last"):
+    def get_close(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        method: str = "ts_data_last",
+    ) -> Union[None, int, float, bool, IndexData]:
         return self.quote.get_data(stock_id, start_time, end_time, field="$close", method=method)
 
-    def get_volume(self, stock_id, start_time, end_time, method="sum"):
+    def get_volume(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        method: Optional[str] = "sum",
+    ) -> Union[None, int, float, bool, IndexData]:
         """get the total deal volume of stock with `stock_id` between the time interval [start_time, end_time)"""
         return self.quote.get_data(stock_id, start_time, end_time, field="$volume", method=method)
 
-    def get_deal_price(self, stock_id, start_time, end_time, direction: OrderDir, method="ts_data_last"):
+    def get_deal_price(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        direction: OrderDir,
+        method: Optional[str] = "ts_data_last",
+    ) -> Union[None, int, float, bool, IndexData]:
         if direction == OrderDir.SELL:
             pstr = self.sell_price
         elif direction == OrderDir.BUY:
             pstr = self.buy_price
         else:
             raise NotImplementedError(f"This type of input is not supported")
+
         deal_price = self.quote.get_data(stock_id, start_time, end_time, field=pstr, method=method)
         if method is not None and (deal_price is None or np.isnan(deal_price) or deal_price <= 1e-08):
             self.logger.warning(f"(stock_id:{stock_id}, trade_time:{(start_time, end_time)}, {pstr}): {deal_price}!!!")
@@ -420,11 +516,16 @@ class Exchange:
             deal_price = self.get_close(stock_id, start_time, end_time, method)
         return deal_price
 
-    def get_factor(self, stock_id, start_time, end_time) -> Union[float, None]:
+    def get_factor(
+        self,
+        stock_id: str,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+    ) -> Optional[float]:
         """
         Returns
         -------
-        Union[float, None]:
+        Optional[float]:
             `None`: if the stock is suspended `None` may be returned
             `float`: return factor if the factor exists
         """
@@ -434,11 +535,16 @@ class Exchange:
         return self.quote.get_data(stock_id, start_time, end_time, field="$factor", method="ts_data_last")
 
     def generate_amount_position_from_weight_position(
-        self, weight_position, cash, start_time, end_time, direction=OrderDir.BUY
-    ):
+        self,
+        weight_position: dict,
+        cash: float,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        direction: OrderDir = OrderDir.BUY,
+    ) -> dict:
         """
-        The generate the target position according to the weight and the cash.
-        NOTE: All the cash will assigned to the tadable stock.
+        Generates the target position according to the weight and the cash.
+        NOTE: All the cash will be assigned to the tradable stock.
         Parameter:
         weight_position : dict {stock_id : weight}; allocate cash by weight_position
             among then, weight must be in this range: 0 < weight < 1
@@ -451,15 +557,14 @@ class Exchange:
 
         # calculate the total weight of tradable value
         tradable_weight = 0.0
-        for stock_id in weight_position:
+        for stock_id, wp in weight_position.items():
             if self.is_stock_tradable(stock_id=stock_id, start_time=start_time, end_time=end_time):
                 # weight_position must be greater than 0 and less than 1
-                if weight_position[stock_id] < 0 or weight_position[stock_id] > 1:
+                if wp < 0 or wp > 1:
                     raise ValueError(
-                        "weight_position is {}, "
-                        "weight_position is not in the range of (0, 1).".format(weight_position[stock_id])
+                        "weight_position is {}, " "weight_position is not in the range of (0, 1).".format(wp),
                     )
-                tradable_weight += weight_position[stock_id]
+                tradable_weight += wp
 
         if tradable_weight - 1.0 >= 1e-5:
             raise ValueError("tradable_weight is {}, can not greater than 1.".format(tradable_weight))
@@ -467,19 +572,24 @@ class Exchange:
         amount_dict = {}
         for stock_id in weight_position:
             if weight_position[stock_id] > 0.0 and self.is_stock_tradable(
-                stock_id=stock_id, start_time=start_time, end_time=end_time
+                stock_id=stock_id,
+                start_time=start_time,
+                end_time=end_time,
             ):
                 amount_dict[stock_id] = (
                     cash
                     * weight_position[stock_id]
                     / tradable_weight
                     // self.get_deal_price(
-                        stock_id=stock_id, start_time=start_time, end_time=end_time, direction=direction
+                        stock_id=stock_id,
+                        start_time=start_time,
+                        end_time=end_time,
+                        direction=direction,
                     )
                 )
         return amount_dict
 
-    def get_real_deal_amount(self, current_amount, target_amount, factor):
+    def get_real_deal_amount(self, current_amount: float, target_amount: float, factor: float | None = None) -> float:
         """
         Calculate the real adjust deal amount when considering the trading unit
         :param current_amount:
@@ -501,7 +611,13 @@ class Exchange:
                 deal_amount = self.round_amount_by_trade_unit(deal_amount, factor)
                 return -deal_amount
 
-    def generate_order_for_target_amount_position(self, target_position, current_position, start_time, end_time):
+    def generate_order_for_target_amount_position(
+        self,
+        target_position: dict,
+        current_position: dict,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+    ) -> List[Order]:
         """
         Note: some future information is used in this function
         Parameter:
@@ -517,7 +633,8 @@ class Exchange:
         # three parts: kept stock_id, dropped stock_id, new stock_id
         # handle kept stock_id
 
-        # because the order of the set is not fixed, the trading order of the stock is different, so that the backtest results of the same parameter are different;
+        # because the order of the set is not fixed, the trading order of the stock is different, so that the backtest
+        # results of the same parameter are different;
         # so here we sort stock_id, and then randomly shuffle the order of stock_id
         # because the same random seed is used, the final stock_id order is fixed
         sorted_ids = sorted(set(list(current_position.keys()) + list(target_position.keys())))
@@ -525,7 +642,7 @@ class Exchange:
         random.shuffle(sorted_ids)
         for stock_id in sorted_ids:
 
-            # Do not generate order for the nontradable stocks
+            # Do not generate order for the non-tradable stocks
             if not self.is_stock_tradable(stock_id=stock_id, start_time=start_time, end_time=end_time):
                 continue
 
@@ -546,7 +663,7 @@ class Exchange:
                         start_time=start_time,
                         end_time=end_time,
                         factor=factor,
-                    )
+                    ),
                 )
             else:
                 # sell stock
@@ -558,14 +675,19 @@ class Exchange:
                         start_time=start_time,
                         end_time=end_time,
                         factor=factor,
-                    )
+                    ),
                 )
         # return order_list : buy + sell
         return sell_order_list + buy_order_list
 
     def calculate_amount_position_value(
-        self, amount_dict, start_time, end_time, only_tradable=False, direction=OrderDir.SELL
-    ):
+        self,
+        amount_dict: dict,
+        start_time: pd.Timestamp,
+        end_time: pd.Timestamp,
+        only_tradable: bool = False,
+        direction: OrderDir = OrderDir.SELL,
+    ) -> float:
         """Parameter
         position : Position()
         amount_dict : {stock_id : amount}
@@ -576,30 +698,44 @@ class Exchange:
         """
         value = 0
         for stock_id in amount_dict:
-            if (
-                only_tradable is True
-                and self.check_stock_suspended(stock_id=stock_id, start_time=start_time, end_time=end_time) is False
-                and self.check_stock_limit(stock_id=stock_id, start_time=start_time, end_time=end_time) is False
-                or only_tradable is False
+            if not only_tradable or (
+                not self.check_stock_suspended(stock_id=stock_id, start_time=start_time, end_time=end_time)
+                and not self.check_stock_limit(stock_id=stock_id, start_time=start_time, end_time=end_time)
             ):
                 value += (
                     self.get_deal_price(
-                        stock_id=stock_id, start_time=start_time, end_time=end_time, direction=direction
+                        stock_id=stock_id,
+                        start_time=start_time,
+                        end_time=end_time,
+                        direction=direction,
                     )
                     * amount_dict[stock_id]
                 )
         return value
 
-    def _get_factor_or_raise_error(self, factor: float = None, stock_id: str = None, start_time=None, end_time=None):
+    def _get_factor_or_raise_error(
+        self,
+        factor: float | None = None,
+        stock_id: str | None = None,
+        start_time: pd.Timestamp = None,
+        end_time: pd.Timestamp = None,
+    ) -> float:
         """Please refer to the docs of get_amount_of_trade_unit"""
         if factor is None:
             if stock_id is not None and start_time is not None and end_time is not None:
                 factor = self.get_factor(stock_id=stock_id, start_time=start_time, end_time=end_time)
             else:
                 raise ValueError(f"`factor` and (`stock_id`, `start_time`, `end_time`) can't both be None")
+        assert factor is not None
         return factor
 
-    def get_amount_of_trade_unit(self, factor: float = None, stock_id: str = None, start_time=None, end_time=None):
+    def get_amount_of_trade_unit(
+        self,
+        factor: float | None = None,
+        stock_id: str | None = None,
+        start_time: pd.Timestamp = None,
+        end_time: pd.Timestamp = None,
+    ) -> Optional[float]:
         """
         get the trade unit of amount based on **factor**
         the factor can be given directly or calculated in given time range and stock id.
@@ -617,15 +753,23 @@ class Exchange:
         """
         if not self.trade_w_adj_price and self.trade_unit is not None:
             factor = self._get_factor_or_raise_error(
-                factor=factor, stock_id=stock_id, start_time=start_time, end_time=end_time
+                factor=factor,
+                stock_id=stock_id,
+                start_time=start_time,
+                end_time=end_time,
             )
             return self.trade_unit / factor
         else:
             return None
 
     def round_amount_by_trade_unit(
-        self, deal_amount, factor: float = None, stock_id: str = None, start_time=None, end_time=None
-    ):
+        self,
+        deal_amount: float,
+        factor: float | None = None,
+        stock_id: str | None = None,
+        start_time: pd.Timestamp = None,
+        end_time: pd.Timestamp = None,
+    ) -> float:
         """Parameter
         Please refer to the docs of get_amount_of_trade_unit
         deal_amount : float, adjusted amount
@@ -635,12 +779,15 @@ class Exchange:
         if not self.trade_w_adj_price and self.trade_unit is not None:
             # the minimal amount is 1. Add 0.1 for solving precision problem.
             factor = self._get_factor_or_raise_error(
-                factor=factor, stock_id=stock_id, start_time=start_time, end_time=end_time
+                factor=factor,
+                stock_id=stock_id,
+                start_time=start_time,
+                end_time=end_time,
             )
             return (deal_amount * factor + 0.1) // self.trade_unit * self.trade_unit / factor
         return deal_amount
 
-    def _clip_amount_by_volume(self, order: Order, dealt_order_amount: dict) -> int:
+    def _clip_amount_by_volume(self, order: Order, dealt_order_amount: dict) -> Optional[float]:
         """parse the capacity limit string and return the actual amount of orders that can be executed.
         NOTE:
             this function will change the order.deal_amount **inplace**
@@ -652,17 +799,14 @@ class Exchange:
         dealt_order_amount : dict
             :param dealt_order_amount: the dealt order amount dict with the format of {stock_id: float}
         """
-        if order.direction == Order.BUY:
-            vol_limit = self.buy_vol_limit
-        elif order.direction == Order.SELL:
-            vol_limit = self.sell_vol_limit
+        vol_limit = self.buy_vol_limit if order.direction == Order.BUY else self.sell_vol_limit
 
         if vol_limit is None:
             return order.deal_amount
 
-        vol_limit_num = []
+        vol_limit_num: List[float] = []
         for limit in vol_limit:
-            assert isinstance(limit, tuple)
+            assert isinstance(limit, tuple) or (isinstance(limit, list) and len(limit) == 2)
             if limit[0] == "current":
                 limit_value = self.quote.get_data(
                     order.stock_id,
@@ -671,7 +815,7 @@ class Exchange:
                     field=limit[1],
                     method="sum",
                 )
-                vol_limit_num.append(limit_value)
+                vol_limit_num.append(cast(float, limit_value))
             elif limit[0] == "cum":
                 limit_value = self.quote.get_data(
                     order.stock_id,
@@ -689,12 +833,14 @@ class Exchange:
         if vol_limit_min < orig_deal_amount:
             self.logger.debug(f"Order clipped due to volume limitation: {order}, {list(zip(vol_limit_num, vol_limit))}")
 
-    def _get_buy_amount_by_cash_limit(self, trade_price, cash, cost_ratio):
+        return None
+
+    def _get_buy_amount_by_cash_limit(self, trade_price: float, cash: float, cost_ratio: float) -> float:
         """return the real order amount after cash limit for buying.
         Parameters
         ----------
         trade_price : float
-        position : cash
+        cash : float
         cost_ratio : float
 
         Return
@@ -702,7 +848,7 @@ class Exchange:
         float
             the real order amount after cash limit for buying.
         """
-        max_trade_amount = 0
+        max_trade_amount = 0.0
         if cash >= self.min_cost:
             # critical_price means the stock transaction price when the service fee is equal to min_cost.
             critical_price = self.min_cost / cost_ratio + self.min_cost
@@ -714,7 +860,12 @@ class Exchange:
                 max_trade_amount = (cash - self.min_cost) / trade_price
         return max_trade_amount
 
-    def _calc_trade_info_by_order(self, order, position: Position, dealt_order_amount):
+    def _calc_trade_info_by_order(
+        self,
+        order: Order,
+        position: Optional[BasePosition],
+        dealt_order_amount: dict,
+    ) -> Tuple[float, float, float]:
         """
         Calculation of trade info
         **NOTE**: Order will be changed in this function
@@ -723,8 +874,11 @@ class Exchange:
         :param dealt_order_amount: the dealt order amount dict with the format of {stock_id: float}
         :return: trade_price, trade_val, trade_cost
         """
-        trade_price = self.get_deal_price(order.stock_id, order.start_time, order.end_time, direction=order.direction)
-        total_trade_val = self.get_volume(order.stock_id, order.start_time, order.end_time) * trade_price
+        trade_price = cast(
+            float,
+            self.get_deal_price(order.stock_id, order.start_time, order.end_time, direction=order.direction),
+        )
+        total_trade_val = cast(float, self.get_volume(order.stock_id, order.start_time, order.end_time)) * trade_price
         order.factor = self.get_factor(order.stock_id, order.start_time, order.end_time)
         order.deal_amount = order.amount  # set to full amount and clip it step by step
         # Clipping amount first
@@ -753,7 +907,8 @@ class Exchange:
                 if not np.isclose(order.deal_amount, current_amount):
                     # when not selling last stock. rounding is necessary
                     order.deal_amount = self.round_amount_by_trade_unit(
-                        min(current_amount, order.deal_amount), order.factor
+                        min(current_amount, order.deal_amount),
+                        order.factor,
                     )
 
                 # in case of negative value of cash
@@ -778,7 +933,8 @@ class Exchange:
                     # The money is not enough
                     max_buy_amount = self._get_buy_amount_by_cash_limit(trade_price, cash, cost_ratio)
                     order.deal_amount = self.round_amount_by_trade_unit(
-                        min(max_buy_amount, order.deal_amount), order.factor
+                        min(max_buy_amount, order.deal_amount),
+                        order.factor,
                     )
                     self.logger.debug(f"Order clipped due to cash limitation: {order}")
                 else:
@@ -789,7 +945,7 @@ class Exchange:
                 order.deal_amount = self.round_amount_by_trade_unit(order.deal_amount, order.factor)
 
         else:
-            raise NotImplementedError("order type {} error".format(order.type))
+            raise NotImplementedError("order direction {} error".format(order.direction))
 
         trade_val = order.deal_amount * trade_price
         trade_cost = max(trade_val * cost_ratio, self.min_cost)

qlib/backtest/executor.py

@@ -1,19 +1,22 @@
-from abc import abstractmethod
+from __future__ import annotations
+
 import copy
+from abc import abstractmethod
+from collections import defaultdict
+from types import GeneratorType
+from typing import Any, Dict, Generator, List, Tuple, Union, cast
+
+import pandas as pd
+
+from qlib.backtest.account import Account
 from qlib.backtest.position import BasePosition
 from qlib.log import get_module_logger
-from types import GeneratorType
-from qlib.backtest.account import Account
-import pandas as pd
-from typing import List, Tuple, Union
-from collections import defaultdict
 
-from .decision import Order, BaseTradeDecision
-from .exchange import Exchange
-from .utils import TradeCalendarManager, CommonInfrastructure, LevelInfrastructure, get_start_end_idx
-
-from ..utils import init_instance_by_config
 from ..strategy.base import BaseStrategy
+from ..utils import init_instance_by_config
+from .decision import BaseTradeDecision, Order
+from .exchange import Exchange
+from .utils import CommonInfrastructure, LevelInfrastructure, TradeCalendarManager, get_start_end_idx
 
 
 class BaseExecutor:
@@ -28,11 +31,11 @@ class BaseExecutor:
         generate_portfolio_metrics: bool = False,
         verbose: bool = False,
         track_data: bool = False,
-        trade_exchange: Exchange = None,
-        common_infra: CommonInfrastructure = None,
-        settle_type=BasePosition.ST_NO,
-        **kwargs,
-    ):
+        trade_exchange: Exchange | None = None,
+        common_infra: CommonInfrastructure | None = None,
+        settle_type: str = BasePosition.ST_NO,
+        **kwargs: Any,
+    ) -> None:
         """
         Parameters
         ----------
@@ -53,15 +56,21 @@ class BaseExecutor:
                 - 'base_price': the based price than which the trading price is advanced, Optional, default by 'twap'
                     - If 'base_price' is 'twap', the based price is the time weighted average price
                     - If 'base_price' is 'vwap', the based price is the volume weighted average price
-                - 'weight_method': weighted method when calculating total trading pa by different orders' pa in each step, optional, default by 'mean'
+                - 'weight_method': weighted method when calculating total trading pa by different orders' pa in each
+                    step, optional, default by 'mean'
                     - If 'weight_method' is 'mean', calculating mean value of different orders' pa
-                    - If 'weight_method' is 'amount_weighted', calculating amount weighted average value of different orders' pa
-                    - If 'weight_method' is 'value_weighted', calculating value weighted average value of different orders' pa
+                    - If 'weight_method' is 'amount_weighted', calculating amount weighted average value of different
+                        orders' pa
+                    - If 'weight_method' is 'value_weighted', calculating value weighted average value of different
+                        orders' pa
             - 'ffr_config': config for calculating fulfill rate(ffr), optional
-                - 'weight_method': weighted method when calculating total trading ffr by different orders' ffr in each step, optional, default by 'mean'
+                - 'weight_method': weighted method when calculating total trading ffr by different orders' ffr in each
+                    step, optional, default by 'mean'
                     - If 'weight_method' is 'mean', calculating mean value of different orders' ffr
-                    - If 'weight_method' is 'amount_weighted', calculating amount weighted average value of different orders' ffr
-                    - If 'weight_method' is 'value_weighted', calculating value weighted average value of different orders' ffr
+                    - If 'weight_method' is 'amount_weighted', calculating amount weighted average value of different
+                        orders' ffr
+                    - If 'weight_method' is 'value_weighted', calculating value weighted average value of different
+                        orders' ffr
             Example:
                 {
                     'show_indicator': True,
@@ -79,7 +88,8 @@ class BaseExecutor:
             whether to print trading info, by default False
         track_data : bool, optional
             whether to generate trade_decision, will be used when training rl agent
-            - If `self.track_data` is true, when making data for training, the input `trade_decision` of `execute` will be generated by `collect_data`
+            - If `self.track_data` is true, when making data for training, the input `trade_decision` of `execute` will
+                be generated by `collect_data`
             - Else,  `trade_decision` will not be generated
 
         trade_exchange : Exchange
@@ -104,17 +114,17 @@ class BaseExecutor:
         self.track_data = track_data
         self._trade_exchange = trade_exchange
         self.level_infra = LevelInfrastructure()
-        self.level_infra.reset_infra(common_infra=common_infra)
+        self.level_infra.reset_infra(common_infra=common_infra, executor=self)
         self._settle_type = settle_type
         self.reset(start_time=start_time, end_time=end_time, common_infra=common_infra)
         if common_infra is None:
             get_module_logger("BaseExecutor").warning(f"`common_infra` is not set for {self}")
 
         # record deal order amount in one day
-        self.dealt_order_amount = defaultdict(float)
+        self.dealt_order_amount: Dict[str, float] = defaultdict(float)
         self.deal_day = None
 
-    def reset_common_infra(self, common_infra, copy_trade_account=False):
+    def reset_common_infra(self, common_infra: CommonInfrastructure, copy_trade_account: bool = False) -> None:
         """
         reset infrastructure for trading
             - reset trade_account
@@ -124,15 +134,18 @@ class BaseExecutor:
         else:
             self.common_infra.update(common_infra)
 
+        self.level_infra.reset_infra(common_infra=self.common_infra)
+
         if common_infra.has("trade_account"):
-            if copy_trade_account:
-                # NOTE: there is a trick in the code.
-                # shallow copy is used instead of deepcopy.
-                # 1. So positions are shared
-                # 2. Others are not shared, so each level has it own metrics (portfolio and trading metrics)
-                self.trade_account: Account = copy.copy(common_infra.get("trade_account"))
-            else:
-                self.trade_account = common_infra.get("trade_account")
+            # NOTE: there is a trick in the code.
+            # shallow copy is used instead of deepcopy.
+            # 1. So positions are shared
+            # 2. Others are not shared, so each level has it own metrics (portfolio and trading metrics)
+            self.trade_account: Account = (
+                copy.copy(common_infra.get("trade_account"))
+                if copy_trade_account
+                else common_infra.get("trade_account")
+            )
             self.trade_account.reset(freq=self.time_per_step, port_metr_enabled=self.generate_portfolio_metrics)
 
     @property
@@ -148,7 +161,7 @@ class BaseExecutor:
         """
         return self.level_infra.get("trade_calendar")
 
-    def reset(self, common_infra: CommonInfrastructure = None, **kwargs):
+    def reset(self, common_infra: CommonInfrastructure | None = None, **kwargs: Any) -> None:
         """
         - reset `start_time` and `end_time`, used in trade calendar
         - reset `common_infra`, used to reset `trade_account`, `trade_exchange`, .etc
@@ -161,13 +174,13 @@ class BaseExecutor:
         if common_infra is not None:
             self.reset_common_infra(common_infra)
 
-    def get_level_infra(self):
+    def get_level_infra(self) -> LevelInfrastructure:
         return self.level_infra
 
-    def finished(self):
+    def finished(self) -> bool:
         return self.trade_calendar.finished()
 
-    def execute(self, trade_decision: BaseTradeDecision, level: int = 0):
+    def execute(self, trade_decision: BaseTradeDecision, level: int = 0) -> List[object]:
         """execute the trade decision and return the executed result
 
         NOTE: this function is never used directly in the framework. Should we delete it?
@@ -184,14 +197,17 @@ class BaseExecutor:
         execute_result : List[object]
             the executed result for trade decision
         """
-        return_value = {}
+        return_value: dict = {}
         for _decision in self.collect_data(trade_decision, return_value=return_value, level=level):
             pass
-        return return_value.get("execute_result")
+        return cast(list, return_value.get("execute_result"))
 
-    @classmethod
     @abstractmethod
-    def _collect_data(cls, trade_decision: BaseTradeDecision, level: int = 0) -> Tuple[List[object], dict]:
+    def _collect_data(
+        self,
+        trade_decision: BaseTradeDecision,
+        level: int = 0,
+    ) -> Union[Generator[Any, Any, Tuple[List[object], dict]], Tuple[List[object], dict]]:
         """
         Please refer to the doc of collect_data
         The only difference between `_collect_data` and `collect_data` is that some common steps are moved into
@@ -209,8 +225,11 @@ class BaseExecutor:
         """
 
     def collect_data(
-        self, trade_decision: BaseTradeDecision, return_value: dict = None, level: int = 0
-    ) -> List[object]:
+        self,
+        trade_decision: BaseTradeDecision,
+        return_value: dict | None = None,
+        level: int = 0,
+    ) -> Generator[Any, Any, List[object]]:
         """Generator for collecting the trade decision data for rl training
 
         his function will make a step forward
@@ -239,6 +258,7 @@ class BaseExecutor:
         object
             trade decision
         """
+
         if self.track_data:
             yield trade_decision
 
@@ -253,7 +273,9 @@ class BaseExecutor:
         obj = self._collect_data(trade_decision=trade_decision, level=level)
 
         if isinstance(obj, GeneratorType):
-            res, kwargs = yield from obj
+            yield_res = yield from obj
+            assert isinstance(yield_res, tuple) and len(yield_res) == 2
+            res, kwargs = yield_res
         else:
             # Some concrete executor don't have inner decisions
             res, kwargs = obj
@@ -277,9 +299,10 @@ class BaseExecutor:
 
         if return_value is not None:
             return_value.update({"execute_result": res})
+
         return res
 
-    def get_all_executors(self):
+    def get_all_executors(self) -> List[BaseExecutor]:
         """get all executors"""
         return [self]
 
@@ -287,7 +310,8 @@ class BaseExecutor:
 class NestedExecutor(BaseExecutor):
     """
     Nested Executor with inner strategy and executor
-    - At each time `execute` is called, it will call the inner strategy and executor to execute the `trade_decision` in a higher frequency env.
+    - At each time `execute` is called, it will call the inner strategy and executor to execute the `trade_decision`
+        in a higher frequency env.
     """
 
     def __init__(
@@ -303,9 +327,9 @@ class NestedExecutor(BaseExecutor):
         track_data: bool = False,
         skip_empty_decision: bool = True,
         align_range_limit: bool = True,
-        common_infra: CommonInfrastructure = None,
-        **kwargs,
-    ):
+        common_infra: CommonInfrastructure | None = None,
+        **kwargs: Any,
+    ) -> None:
         """
         Parameters
         ----------
@@ -323,10 +347,14 @@ class NestedExecutor(BaseExecutor):
             It is only for nested executor, because range_limit is given by outer strategy
         """
         self.inner_executor: BaseExecutor = init_instance_by_config(
-            inner_executor, common_infra=common_infra, accept_types=BaseExecutor
+            inner_executor,
+            common_infra=common_infra,
+            accept_types=BaseExecutor,
         )
         self.inner_strategy: BaseStrategy = init_instance_by_config(
-            inner_strategy, common_infra=common_infra, accept_types=BaseStrategy
+            inner_strategy,
+            common_infra=common_infra,
+            accept_types=BaseStrategy,
         )
 
         self._skip_empty_decision = skip_empty_decision
@@ -344,10 +372,10 @@ class NestedExecutor(BaseExecutor):
             **kwargs,
         )
 
-    def reset_common_infra(self, common_infra, copy_trade_account=False):
+    def reset_common_infra(self, common_infra: CommonInfrastructure, copy_trade_account: bool = False) -> None:
         """
         reset infrastructure for trading
-            - reset inner_strategyand inner_executor common infra
+            - reset inner_strategy and inner_executor common infra
         """
         # NOTE: please refer to the docs of BaseExecutor.reset_common_infra for the meaning of `copy_trade_account`
 
@@ -358,7 +386,7 @@ class NestedExecutor(BaseExecutor):
         self.inner_executor.reset_common_infra(common_infra, copy_trade_account=True)
         self.inner_strategy.reset_common_infra(common_infra)
 
-    def _init_sub_trading(self, trade_decision):
+    def _init_sub_trading(self, trade_decision: BaseTradeDecision) -> None:
         trade_start_time, trade_end_time = self.trade_calendar.get_step_time()
         self.inner_executor.reset(start_time=trade_start_time, end_time=trade_end_time)
         sub_level_infra = self.inner_executor.get_level_infra()
@@ -368,14 +396,18 @@ class NestedExecutor(BaseExecutor):
     def _update_trade_decision(self, trade_decision: BaseTradeDecision) -> BaseTradeDecision:
         # outer strategy have chance to update decision each iterator
         updated_trade_decision = trade_decision.update(self.inner_executor.trade_calendar)
-        if updated_trade_decision is not None:
+        if updated_trade_decision is not None:  # TODO: always is None for now?
             trade_decision = updated_trade_decision
             # NEW UPDATE
             # create a hook for inner strategy to update outer decision
-            self.inner_strategy.alter_outer_trade_decision(trade_decision)
+            trade_decision = self.inner_strategy.alter_outer_trade_decision(trade_decision)
         return trade_decision
 
-    def _collect_data(self, trade_decision: BaseTradeDecision, level: int = 0):
+    def _collect_data(
+        self,
+        trade_decision: BaseTradeDecision,
+        level: int = 0,
+    ) -> Generator[Any, Any, Tuple[List[object], dict]]:
         execute_result = []
         inner_order_indicators = []
         decision_list = []
@@ -390,8 +422,8 @@ class NestedExecutor(BaseExecutor):
 
             if trade_decision.empty() and self._skip_empty_decision:
                 # give one chance for outer strategy to update the strategy
-                # - For updating some information in the sub executor(the strategy have no knowledge of the inner
-                # executor when generating the decision)
+                # - For updating some information in the sub executor (the strategy have no knowledge of the inner
+                #   executor when generating the decision)
                 break
 
             sub_cal: TradeCalendarManager = self.inner_executor.trade_calendar
@@ -405,15 +437,19 @@ class NestedExecutor(BaseExecutor):
 
                 # NOTE: !!!!!
                 # the two lines below is for a special case in RL
-                # To solve the confliction below
-                # - Normally, user will create a strategy and embed it into Qlib's executor and simulator interaction loop
-                #   For a _nested qlib example_, (Qlib Strategy) <=> (Qlib Executor[(inner Qlib Strategy) <=> (inner Qlib Executor)])
+                # To solve the conflicts below
+                # - Normally, user will create a strategy and embed it into Qlib's executor and simulator interaction
+                #   loop For a _nested qlib example_, (Qlib Strategy) <=> (Qlib Executor[(inner Qlib Strategy) <=>
+                #   (inner Qlib Executor)])
                 # - However, RL-based framework has it's own script to run the loop
                 #   For an _RL learning example_, (RL Policy) <=> (RL Env[(inner Qlib Executor)])
-                # To make it possible to run  _nested qlib example_ and _RL learning example_ together, the solution below is proposed
-                # - The entry script follow the example of  _RL learning example_ to be compatible with all kinds of RL Framework
+                # To make it possible to run  _nested qlib example_ and _RL learning example_ together, the solution
+                # below is proposed
+                # - The entry script follow the example of  _RL learning example_ to be compatible with all kinds of
+                #   RL Framework
                 # - Each step of (RL Env) will make (inner Qlib Executor) one step forward
-                #     - (inner Qlib Strategy) is a proxy strategy, it will give the program control right to (RL Env) by `yield from` and wait for the action from the policy
+                #     - (inner Qlib Strategy) is a proxy strategy, it will give the program control right to (RL Env)
+                #       by `yield from` and wait for the action from the policy
                 # So the two lines below is the implementation of yielding control rights
                 if isinstance(res, GeneratorType):
                     res = yield from res
@@ -427,21 +463,26 @@ class NestedExecutor(BaseExecutor):
 
                 # NOTE: Trade Calendar will step forward in the follow line
                 _inner_execute_result = yield from self.inner_executor.collect_data(
-                    trade_decision=_inner_trade_decision, level=level + 1
+                    trade_decision=_inner_trade_decision,
+                    level=level + 1,
                 )
+                assert isinstance(_inner_execute_result, list)
                 self.post_inner_exe_step(_inner_execute_result)
                 execute_result.extend(_inner_execute_result)
 
                 inner_order_indicators.append(
-                    self.inner_executor.trade_account.get_trade_indicator().get_order_indicator(raw=True)
+                    self.inner_executor.trade_account.get_trade_indicator().get_order_indicator(raw=True),
                 )
             else:
                 # do nothing and just step forward
                 sub_cal.step()
 
+        # Let inner strategy know that the outer level execution is done.
+        self.inner_strategy.post_upper_level_exe_step()
+
         return execute_result, {"inner_order_indicators": inner_order_indicators, "decision_list": decision_list}
 
-    def post_inner_exe_step(self, inner_exe_res):
+    def post_inner_exe_step(self, inner_exe_res: List[object]) -> None:
         """
         A hook for doing sth after each step of inner strategy
 
@@ -450,12 +491,25 @@ class NestedExecutor(BaseExecutor):
         inner_exe_res :
             the execution result of inner task
         """
+        self.inner_strategy.post_exe_step(inner_exe_res)
 
-    def get_all_executors(self):
+    def get_all_executors(self) -> List[BaseExecutor]:
         """get all executors, including self and inner_executor.get_all_executors()"""
         return [self, *self.inner_executor.get_all_executors()]
 
 
+def _retrieve_orders_from_decision(trade_decision: BaseTradeDecision) -> List[Order]:
+    """
+    IDE-friendly helper function.
+    """
+    decisions = trade_decision.get_decision()
+    orders: List[Order] = []
+    for decision in decisions:
+        assert isinstance(decision, Order)
+        orders.append(decision)
+    return orders
+
+
 class SimulatorExecutor(BaseExecutor):
     """Executor that simulate the true market"""
 
@@ -464,10 +518,10 @@ class SimulatorExecutor(BaseExecutor):
 
     # available trade_types
     TT_SERIAL = "serial"
-    ## The orders will be executed serially in a sequence
+    # The orders will be executed serially in a sequence
     # In each trading step, it is possible that users sell instruments first and use the money to buy new instruments
     TT_PARAL = "parallel"
-    ## The orders will be executed parallelly
+    # The orders will be executed in parallel
     # In each trading step, if users try to sell instruments first and buy new instruments with money, failure will
     # occur
 
@@ -480,10 +534,10 @@ class SimulatorExecutor(BaseExecutor):
         generate_portfolio_metrics: bool = False,
         verbose: bool = False,
         track_data: bool = False,
-        common_infra: CommonInfrastructure = None,
+        common_infra: CommonInfrastructure | None = None,
         trade_type: str = TT_SERIAL,
-        **kwargs,
-    ):
+        **kwargs: Any,
+    ) -> None:
         """
         Parameters
         ----------
@@ -517,7 +571,7 @@ class SimulatorExecutor(BaseExecutor):
         List[Order]:
             get a list orders according to `self.trade_type`
         """
-        orders = trade_decision.get_decision()
+        orders = _retrieve_orders_from_decision(trade_decision)
 
         if self.trade_type == self.TT_SERIAL:
             # Orders will be traded in a parallel way
@@ -525,29 +579,26 @@ class SimulatorExecutor(BaseExecutor):
         elif self.trade_type == self.TT_PARAL:
             # NOTE: !!!!!!!
             # Assumption: there will not be orders in different trading direction in a single step of a strategy !!!!
-            # The parallel trading failure will be caused only by the confliction of money
-            # Therefore, make the buying go first will make sure the confliction happen.
+            # The parallel trading failure will be caused only by the conflicts of money
+            # Therefore, make the buying go first will make sure the conflicts happen.
             # It equals to parallel trading after sorting the order by direction
             order_it = sorted(orders, key=lambda order: -order.direction)
         else:
             raise NotImplementedError(f"This type of input is not supported")
         return order_it
 
-    def _update_dealt_order_amount(self, order):
-        """update date and dealt order amount in the day."""
-
-        now_deal_day = self.trade_calendar.get_step_time()[0].floor(freq="D")
-        if self.deal_day is None or now_deal_day > self.deal_day:
-            self.dealt_order_amount = defaultdict(float)
-            self.deal_day = now_deal_day
-        self.dealt_order_amount[order.stock_id] += order.deal_amount
-
-    def _collect_data(self, trade_decision: BaseTradeDecision, level: int = 0):
-
+    def _collect_data(self, trade_decision: BaseTradeDecision, level: int = 0) -> Tuple[List[object], dict]:
         trade_start_time, _ = self.trade_calendar.get_step_time()
-        execute_result = []
+        execute_result: list = []
 
         for order in self._get_order_iterator(trade_decision):
+            # Each time we move into a new date, clear `self.dealt_order_amount` since it only maintains intraday
+            # information.
+            now_deal_day = self.trade_calendar.get_step_time()[0].floor(freq="D")
+            if self.deal_day is None or now_deal_day > self.deal_day:
+                self.dealt_order_amount = defaultdict(float)
+                self.deal_day = now_deal_day
+
             # execute the order.
             # NOTE: The trade_account will be changed in this function
             trade_val, trade_cost, trade_price = self.trade_exchange.deal_order(
@@ -556,10 +607,13 @@ class SimulatorExecutor(BaseExecutor):
                 dealt_order_amount=self.dealt_order_amount,
             )
             execute_result.append((order, trade_val, trade_cost, trade_price))
-            self._update_dealt_order_amount(order)
+
+            self.dealt_order_amount[order.stock_id] += order.deal_amount
+
             if self.verbose:
                 print(
-                    "[I {:%Y-%m-%d %H:%M:%S}]: {} {}, price {:.2f}, amount {}, deal_amount {}, factor {}, value {:.2f}, cash {:.2f}.".format(
+                    "[I {:%Y-%m-%d %H:%M:%S}]: {} {}, price {:.2f}, amount {}, deal_amount {}, factor {}, "
+                    "value {:.2f}, cash {:.2f}.".format(
                         trade_start_time,
                         "sell" if order.direction == Order.SELL else "buy",
                         order.stock_id,
@@ -569,6 +623,6 @@ class SimulatorExecutor(BaseExecutor):
                         order.factor,
                         trade_val,
                         self.trade_account.get_cash(),
-                    )
+                    ),
                 )
         return execute_result, {"trade_info": execute_result}