* Fix the bug of reading NA string as NaN in exists_qlib_data.
* Fix the .gitignore file.
* Update the fix and add some comments.
* format with black
---------
Co-authored-by: Chuan Xu <chuan.xu@sas.com>
Co-authored-by: Linlang Lv (iSoftStone Information) <v-lvlinlang@microsoft.com>
* download orderbook data
* fix CI error
* fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* test fix CI error
* optimize get_data code
* optimize get_data code
* optimize get_data code
* optimize README
---------
Co-authored-by: Linlang <v-linlanglv@microsoft.com>
* df value is set as expected, suppress the warning;
* depress warning with pandas option_context
---------
Co-authored-by: Cadenza-Li <362237642@qq.com>
* Add multi pass port ana record
* Add list function
* Add documentation and support <MODEL> tag
* Add drop in replacement example
* reformat
* Change according to comments
* update format
* Update record_temp.py
Fix type hint
* Update record_temp.py
* Intermediate version
* Fix yaml template & Successfully run rolling
* Be compatible with benchmark
* Get same results with previous linear model
* Black formatting
* Update black
* Update the placeholder mechanism
* Update CI
* Update CI
* Upgrade Black
* Fix CI and simplify code
* Fix CI
* Move the data processing caching mechanism into utils.
* Adjusting DDG-DA
* Organize import
* transpose dimension 1 and 2 to match nn.Conv1d input
* 1.update TCN benchmarks;
2.Emphasize updating the benchmark table;
* replace specific version with main
---------
Co-authored-by: lijinhui <362237642@qq.com>
* Remove lr_decay and lr_decay_steps params
More flexible way to pass a scheduler (via callable function) is already
supported
* remove lr_decay and lr_decay_steps from mlp workflow configs
* wip
* wip
* wip
* Fix naming errors
* Backtest test passed
* Why training stuck?
* Minor
* Refine train configs
* Use dummy in training
* Remove pickle_dataframe
* CI
* CI
* Add more strict condition to filter orders
* Pass test
* Add TODO in example
---------
Co-authored-by: Young <afe.young@gmail.com>
* Waiting for bin data
* Complete readme
* CI
* Add inst filter by time
* Update qlib/data/dataset/processor.py
* typo
* Fix time filter bug
* Add Filter and set Universe
* Complete data pipeline
* Fix Provider Logger Info Args
* Add DQN; a minor bugfix in ppo reward.
* update readme. modify assertion logic in strategy check.
* Fix Doc issues and fix black
* Fix pylint Error
---------
Co-authored-by: Young <afe.young@gmail.com>
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
* Workflow runnable
* CI
* Slight changes to make the workflow runnable. The changes of handler/provider should be reverted before merging.
* Train experiment successful
* Refine handler & provider
* test passed
* Ready to test on server
* Minor
* Test passed
* TWAP training
* Add PPOReward
* Add a FIXME
* Refine PPO reward according to PR comments
* Minor
* Resolve PR comments
* CI issues
* CI issues
* CI issues
* Workflow runnable
* CI
* Slight changes to make the workflow runnable. The changes of handler/provider should be reverted before merging.
* Train experiment successful
* Refine handler & provider
* CI issues
* Resolve PR comments
* Resolve PR comments
* CI issues
* Fix test issue
* Black
* 1) check limit_up/down should consider direction; 2) fix some typo, typehint etc
* fix error
* Update test_all_pipeline.py
Believe it's just some arbitrary number.
The excess return is expected to change when trading logic changes.
* add flag forbid_all_trade_at_limit to keep previous behivour for backward compatibility
* config.py: fix typo; static method
* fix typo in qlib/utils/paral
* 1) limit numpy version as numba support for 1.24+ has not been released; 2) no need to use custom numba version for pytest.
* remove useless argument
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
* horizontally put the bar figures
* 1) use rangebreaks to handle gaps in datetime axis instead of make them string; 2) allow simultaneously plot rankic in ic_figure
* pylint improvement
* fix black lint
* better axis formatting
* default not show gaps
* resolve doc built error
* fix pylint
* Update qlib/contrib/report/analysis_model/analysis_model_performance.py
More detailed description
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
* Update qlib/contrib/report/analysis_model/analysis_model_performance.py
for Python backward compatibility
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
* add doc string
* fix black
* 1) limit numpy version as numba support for 1.24+ has not been released; 2) no need to use custom numba version for pytest.
Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>
