mirror of
https://github.com/microsoft/qlib.git
synced 2026-07-05 20:11:08 +08:00
354 lines
11 KiB
Python
354 lines
11 KiB
Python
# Copyright (c) Microsoft Corporation.
|
|
# Licensed under the MIT License.
|
|
|
|
|
|
from __future__ import division
|
|
from __future__ import print_function
|
|
|
|
import os
|
|
import numpy as np
|
|
import pandas as pd
|
|
import copy
|
|
from sklearn.metrics import roc_auc_score, mean_squared_error
|
|
import logging
|
|
from ...utils import (
|
|
unpack_archive_with_buffer,
|
|
save_multiple_parts_file,
|
|
create_save_path,
|
|
drop_nan_by_y_index,
|
|
)
|
|
from ...log import get_module_logger, TimeInspector
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
import torch.optim as optim
|
|
|
|
from ...model.base import Model
|
|
from ...data.dataset import DatasetH
|
|
from ...data.dataset.handler import DataHandlerLP
|
|
|
|
|
|
class ALSTM(Model):
|
|
"""ALSTM Model
|
|
|
|
Parameters
|
|
----------
|
|
d_feat : int
|
|
input dimension for each time step
|
|
metric: str
|
|
the evaluate metric used in early stop
|
|
optimizer : str
|
|
optimizer name
|
|
GPU : str
|
|
the GPU ID(s) used for training
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
d_feat=6,
|
|
hidden_size=64,
|
|
num_layers=2,
|
|
dropout=0.0,
|
|
n_epochs=200,
|
|
lr=0.001,
|
|
metric="",
|
|
batch_size=2000,
|
|
early_stop=20,
|
|
loss="mse",
|
|
optimizer="adam",
|
|
GPU="0",
|
|
seed=None,
|
|
**kwargs
|
|
):
|
|
# Set logger.
|
|
self.logger = get_module_logger("ALSTM")
|
|
self.logger.info("ALSTM pytorch version...")
|
|
|
|
# set hyper-parameters.
|
|
self.d_feat = d_feat
|
|
self.hidden_size = hidden_size
|
|
self.num_layers = num_layers
|
|
self.dropout = dropout
|
|
self.n_epochs = n_epochs
|
|
self.lr = lr
|
|
self.metric = metric
|
|
self.batch_size = batch_size
|
|
self.early_stop = early_stop
|
|
self.optimizer = optimizer.lower()
|
|
self.loss = loss
|
|
self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() else "cpu")
|
|
self.use_gpu = torch.cuda.is_available()
|
|
self.seed = seed
|
|
|
|
self.logger.info(
|
|
"ALSTM parameters setting:"
|
|
"\nd_feat : {}"
|
|
"\nhidden_size : {}"
|
|
"\nnum_layers : {}"
|
|
"\ndropout : {}"
|
|
"\nn_epochs : {}"
|
|
"\nlr : {}"
|
|
"\nmetric : {}"
|
|
"\nbatch_size : {}"
|
|
"\nearly_stop : {}"
|
|
"\noptimizer : {}"
|
|
"\nloss_type : {}"
|
|
"\nvisible_GPU : {}"
|
|
"\nuse_GPU : {}"
|
|
"\nseed : {}".format(
|
|
d_feat,
|
|
hidden_size,
|
|
num_layers,
|
|
dropout,
|
|
n_epochs,
|
|
lr,
|
|
metric,
|
|
batch_size,
|
|
early_stop,
|
|
optimizer.lower(),
|
|
loss,
|
|
GPU,
|
|
self.use_gpu,
|
|
seed,
|
|
)
|
|
)
|
|
|
|
if self.seed is not None:
|
|
np.random.seed(self.seed)
|
|
torch.manual_seed(self.seed)
|
|
|
|
self.ALSTM_model = ALSTMModel(
|
|
d_feat=self.d_feat,
|
|
hidden_size=self.hidden_size,
|
|
num_layers=self.num_layers,
|
|
dropout=self.dropout,
|
|
)
|
|
if optimizer.lower() == "adam":
|
|
self.train_optimizer = optim.Adam(self.ALSTM_model.parameters(), lr=self.lr)
|
|
elif optimizer.lower() == "gd":
|
|
self.train_optimizer = optim.SGD(self.ALSTM_model.parameters(), lr=self.lr)
|
|
else:
|
|
raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
|
|
|
|
self._fitted = False
|
|
self.ALSTM_model.to(self.device)
|
|
|
|
def mse(self, pred, label):
|
|
loss = (pred - label) ** 2
|
|
return torch.mean(loss)
|
|
|
|
def loss_fn(self, pred, label):
|
|
mask = ~torch.isnan(label)
|
|
|
|
if self.loss == "mse":
|
|
return self.mse(pred[mask], label[mask])
|
|
|
|
raise ValueError("unknown loss `%s`" % self.loss)
|
|
|
|
def metric_fn(self, pred, label):
|
|
|
|
mask = torch.isfinite(label)
|
|
|
|
if self.metric == "" or self.metric == "loss":
|
|
return -self.loss_fn(pred[mask], label[mask])
|
|
|
|
raise ValueError("unknown metric `%s`" % self.metric)
|
|
|
|
def train_epoch(self, x_train, y_train):
|
|
|
|
x_train_values = x_train.values
|
|
y_train_values = np.squeeze(y_train.values)
|
|
|
|
self.ALSTM_model.train()
|
|
|
|
indices = np.arange(len(x_train_values))
|
|
np.random.shuffle(indices)
|
|
|
|
for i in range(len(indices))[:: self.batch_size]:
|
|
|
|
if len(indices) - i < self.batch_size:
|
|
break
|
|
|
|
feature = torch.from_numpy(x_train_values[indices[i : i + self.batch_size]]).float().to(self.device)
|
|
label = torch.from_numpy(y_train_values[indices[i : i + self.batch_size]]).float().to(self.device)
|
|
|
|
pred = self.ALSTM_model(feature)
|
|
loss = self.loss_fn(pred, label)
|
|
|
|
self.train_optimizer.zero_grad()
|
|
loss.backward()
|
|
torch.nn.utils.clip_grad_value_(self.ALSTM_model.parameters(), 3.0)
|
|
self.train_optimizer.step()
|
|
|
|
def test_epoch(self, data_x, data_y):
|
|
|
|
# prepare training data
|
|
x_values = data_x.values
|
|
y_values = np.squeeze(data_y.values)
|
|
|
|
self.ALSTM_model.eval()
|
|
|
|
scores = []
|
|
losses = []
|
|
|
|
indices = np.arange(len(x_values))
|
|
|
|
for i in range(len(indices))[:: self.batch_size]:
|
|
|
|
if len(indices) - i < self.batch_size:
|
|
break
|
|
|
|
feature = torch.from_numpy(x_values[indices[i : i + self.batch_size]]).float().to(self.device)
|
|
label = torch.from_numpy(y_values[indices[i : i + self.batch_size]]).float().to(self.device)
|
|
|
|
pred = self.ALSTM_model(feature)
|
|
loss = self.loss_fn(pred, label)
|
|
losses.append(loss.item())
|
|
|
|
score = self.metric_fn(pred, label)
|
|
scores.append(score.item())
|
|
|
|
return np.mean(losses), np.mean(scores)
|
|
|
|
def fit(
|
|
self,
|
|
dataset: DatasetH,
|
|
evals_result=dict(),
|
|
verbose=True,
|
|
save_path=None,
|
|
):
|
|
|
|
df_train, df_valid, df_test = dataset.prepare(
|
|
["train", "valid", "test"],
|
|
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"]
|
|
|
|
if save_path == None:
|
|
save_path = create_save_path(save_path)
|
|
stop_steps = 0
|
|
train_loss = 0
|
|
best_score = -np.inf
|
|
best_epoch = 0
|
|
evals_result["train"] = []
|
|
evals_result["valid"] = []
|
|
|
|
# train
|
|
self.logger.info("training...")
|
|
self._fitted = True
|
|
|
|
for step in range(self.n_epochs):
|
|
self.logger.info("Epoch%d:", step)
|
|
self.logger.info("training...")
|
|
self.train_epoch(x_train, y_train)
|
|
self.logger.info("evaluating...")
|
|
train_loss, train_score = self.test_epoch(x_train, y_train)
|
|
val_loss, val_score = self.test_epoch(x_valid, y_valid)
|
|
self.logger.info("train %.6f, valid %.6f" % (train_score, val_score))
|
|
evals_result["train"].append(train_score)
|
|
evals_result["valid"].append(val_score)
|
|
|
|
if val_score > best_score:
|
|
best_score = val_score
|
|
stop_steps = 0
|
|
best_epoch = step
|
|
best_param = copy.deepcopy(self.ALSTM_model.state_dict())
|
|
else:
|
|
stop_steps += 1
|
|
if stop_steps >= self.early_stop:
|
|
self.logger.info("early stop")
|
|
break
|
|
|
|
self.logger.info("best score: %.6lf @ %d" % (best_score, best_epoch))
|
|
self.ALSTM_model.load_state_dict(best_param)
|
|
torch.save(best_param, save_path)
|
|
|
|
if self.use_gpu:
|
|
torch.cuda.empty_cache()
|
|
|
|
def predict(self, dataset):
|
|
if not self._fitted:
|
|
raise ValueError("model is not fitted yet!")
|
|
|
|
x_test = dataset.prepare("test", col_set="feature")
|
|
index = x_test.index
|
|
self.ALSTM_model.eval()
|
|
x_values = x_test.values
|
|
sample_num = x_values.shape[0]
|
|
preds = []
|
|
|
|
for begin in range(sample_num)[:: self.batch_size]:
|
|
|
|
if sample_num - begin < self.batch_size:
|
|
end = sample_num
|
|
else:
|
|
end = begin + self.batch_size
|
|
|
|
x_batch = torch.from_numpy(x_values[begin:end]).float().to(self.device)
|
|
|
|
with torch.no_grad():
|
|
if self.use_gpu:
|
|
pred = self.ALSTM_model(x_batch).detach().cpu().numpy()
|
|
else:
|
|
pred = self.ALSTM_model(x_batch).detach().numpy()
|
|
|
|
preds.append(pred)
|
|
|
|
return pd.Series(np.concatenate(preds), index=index)
|
|
|
|
|
|
class ALSTMModel(nn.Module):
|
|
def __init__(self, d_feat=6, hidden_size=64, num_layers=2, dropout=0.0, rnn_type="GRU"):
|
|
super().__init__()
|
|
self.hid_size = hidden_size
|
|
self.input_size = d_feat
|
|
self.dropout = dropout
|
|
self.rnn_type = rnn_type
|
|
self.rnn_layer = num_layers
|
|
self._build_model()
|
|
|
|
def _build_model(self):
|
|
try:
|
|
klass = getattr(nn, self.rnn_type.upper())
|
|
except:
|
|
raise ValueError("unknown rnn_type `%s`" % self.rnn_type)
|
|
self.net = nn.Sequential()
|
|
self.net.add_module("fc_in", nn.Linear(in_features=self.input_size, out_features=self.hid_size))
|
|
self.net.add_module("act", nn.Tanh())
|
|
self.rnn = klass(
|
|
input_size=self.hid_size,
|
|
hidden_size=self.hid_size,
|
|
num_layers=self.rnn_layer,
|
|
batch_first=True,
|
|
dropout=self.dropout,
|
|
)
|
|
self.fc_out = nn.Linear(in_features=self.hid_size * 2, out_features=1)
|
|
self.att_net = nn.Sequential()
|
|
self.att_net.add_module(
|
|
"att_fc_in",
|
|
nn.Linear(in_features=self.hid_size, out_features=int(self.hid_size / 2)),
|
|
)
|
|
self.att_net.add_module("att_dropout", torch.nn.Dropout(self.dropout))
|
|
self.att_net.add_module("att_act", nn.Tanh())
|
|
self.att_net.add_module(
|
|
"att_fc_out",
|
|
nn.Linear(in_features=int(self.hid_size / 2), out_features=1, bias=False),
|
|
)
|
|
self.att_net.add_module("att_softmax", nn.Softmax(dim=1))
|
|
|
|
def forward(self, inputs):
|
|
# inputs: [batch_size, input_size*input_day]
|
|
inputs = inputs.view(len(inputs), self.input_size, -1)
|
|
inputs = inputs.permute(0, 2, 1) # [batch, input_size, seq_len] -> [batch, seq_len, input_size]
|
|
rnn_out, _ = self.rnn(self.net(inputs)) # [batch, seq_len, num_directions * hidden_size]
|
|
attention_score = self.att_net(rnn_out) # [batch, seq_len, 1]
|
|
out_att = torch.mul(rnn_out, attention_score)
|
|
out_att = torch.sum(out_att, dim=1)
|
|
out = self.fc_out(
|
|
torch.cat((rnn_out[:, -1, :], out_att), dim=1)
|
|
) # [batch, seq_len, num_directions * hidden_size] -> [batch, 1]
|
|
return out[..., 0]
|