liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-14 16:26:55 +08:00
Code Activity

Update models to enable save/load

Browse Source
This commit is contained in:
Jactus
2021-02-04 17:23:10 +08:00
committed by you-n-g
parent 6a670828a5
commit b84686b215
12 changed files with 35 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
qlib/contrib/model/pytorch_alstm.py
View File

@@ -130,7 +130,7 @@ class ALSTM(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.ALSTM_model.to(self.device) self.ALSTM_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -238,7 +238,7 @@ class ALSTM(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -270,7 +270,7 @@ class ALSTM(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature") x_test = dataset.prepare("test", col_set="feature")

6
qlib/contrib/model/pytorch_alstm_ts.py
View File

@@ -135,7 +135,7 @@ class ALSTM(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.ALSTM_model.to(self.device) self.ALSTM_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -225,7 +225,7 @@ class ALSTM(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -257,7 +257,7 @@ class ALSTM(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I) dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I)

6
qlib/contrib/model/pytorch_gats.py
View File

@@ -142,7 +142,7 @@ class GATs(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.GAT_model.to(self.device) self.GAT_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -275,7 +275,7 @@ class GATs(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -307,7 +307,7 @@ class GATs(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature") x_test = dataset.prepare("test", col_set="feature")

6
qlib/contrib/model/pytorch_gats_ts.py
View File

@@ -164,7 +164,7 @@ class GATs(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.GAT_model.to(self.device) self.GAT_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -297,7 +297,7 @@ class GATs(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -329,7 +329,7 @@ class GATs(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I) dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I)

6
qlib/contrib/model/pytorch_gru.py
View File

@@ -130,7 +130,7 @@ class GRU(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.gru_model.to(self.device) self.gru_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -238,7 +238,7 @@ class GRU(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -270,7 +270,7 @@ class GRU(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature") x_test = dataset.prepare("test", col_set="feature")

6
qlib/contrib/model/pytorch_gru_ts.py
View File

@@ -135,7 +135,7 @@ class GRU(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.GRU_model.to(self.device) self.GRU_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -225,7 +225,7 @@ class GRU(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -257,7 +257,7 @@ class GRU(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I) dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I)

6
qlib/contrib/model/pytorch_lstm.py
View File

@@ -130,7 +130,7 @@ class LSTM(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.lstm_model.to(self.device) self.lstm_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -238,7 +238,7 @@ class LSTM(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -270,7 +270,7 @@ class LSTM(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature") x_test = dataset.prepare("test", col_set="feature")

6
qlib/contrib/model/pytorch_lstm_ts.py
View File

@@ -135,7 +135,7 @@ class LSTM(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.LSTM_model.to(self.device) self.LSTM_model.to(self.device)
def mse(self, pred, label): def mse(self, pred, label):
@@ -225,7 +225,7 @@ class LSTM(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -257,7 +257,7 @@ class LSTM(Model):
torch.cuda.empty_cache() torch.cuda.empty_cache()
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I) dl_test = dataset.prepare("test", col_set=["feature", "label"], data_key=DataHandlerLP.DK_I)

6
qlib/contrib/model/pytorch_nn.py
View File

@@ -150,7 +150,7 @@ class DNNModelPytorch(Model):
eps=1e-08, eps=1e-08,
) )
self._fitted = False self.fitted = False
self.dnn_model.to(self.device) self.dnn_model.to(self.device)
def fit( def fit(
@@ -180,7 +180,7 @@ class DNNModelPytorch(Model):
evals_result["valid"] = [] evals_result["valid"] = []
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
# return # return
# prepare training data # prepare training data
x_train_values = torch.from_numpy(x_train.values).float() x_train_values = torch.from_numpy(x_train.values).float()
@@ -265,7 +265,7 @@ class DNNModelPytorch(Model):
raise NotImplementedError("loss {} is not supported!".format(loss_type)) raise NotImplementedError("loss {} is not supported!".format(loss_type))
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test_pd = dataset.prepare("test", col_set="feature") x_test_pd = dataset.prepare("test", col_set="feature")
x_test = torch.from_numpy(x_test_pd.values).float().to(self.device) x_test = torch.from_numpy(x_test_pd.values).float().to(self.device)

6
qlib/contrib/model/pytorch_sfm.py
View File

@@ -302,7 +302,7 @@ class SFM(Model):
else: else:
raise NotImplementedError("optimizer {} is not supported!".format(optimizer)) raise NotImplementedError("optimizer {} is not supported!".format(optimizer))
self._fitted = False self.fitted = False
self.sfm_model.to(self.device) self.sfm_model.to(self.device)
def test_epoch(self, data_x, data_y): def test_epoch(self, data_x, data_y):
@@ -386,7 +386,7 @@ class SFM(Model):
# train # train
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for step in range(self.n_epochs): for step in range(self.n_epochs):
self.logger.info("Epoch%d:", step) self.logger.info("Epoch%d:", step)
@@ -435,7 +435,7 @@ class SFM(Model):
raise ValueError("unknown metric `%s`" % self.metric) raise ValueError("unknown metric `%s`" % self.metric)
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature") x_test = dataset.prepare("test", col_set="feature")

5
qlib/contrib/model/pytorch_tabnet.py
View File

@@ -88,6 +88,7 @@ class TabnetModel(Model):
"\nGPU : {}" "\nGPU : {}"
"\npretrain: {}".format(self.batch_size, vbs, GPU, pretrain) "\npretrain: {}".format(self.batch_size, vbs, GPU, pretrain)
) )
self.fitted = False
np.random.seed(self.seed) np.random.seed(self.seed)
torch.manual_seed(self.seed) torch.manual_seed(self.seed)
@@ -187,7 +188,7 @@ class TabnetModel(Model):
evals_result["valid"] = [] evals_result["valid"] = []
self.logger.info("training...") self.logger.info("training...")
self._fitted = True self.fitted = True
for epoch_idx in range(self.n_epochs): for epoch_idx in range(self.n_epochs):
self.logger.info("epoch: %s" % (epoch_idx)) self.logger.info("epoch: %s" % (epoch_idx))
@@ -212,7 +213,7 @@ class TabnetModel(Model):
self.logger.info("best score: %.6lf @ %d" % (best_score, best_epoch)) self.logger.info("best score: %.6lf @ %d" % (best_score, best_epoch))
def predict(self, dataset): def predict(self, dataset):
if not self._fitted: if not self.fitted:
raise ValueError("model is not fitted yet!") raise ValueError("model is not fitted yet!")
x_test = dataset.prepare("test", col_set="feature", data_key=DataHandlerLP.DK_I) x_test = dataset.prepare("test", col_set="feature", data_key=DataHandlerLP.DK_I)

4
qlib/data/data.py
View File

@@ -478,13 +478,13 @@ class DatasetProvider(abc.ABC):
data = pd.DataFrame(obj) data = pd.DataFrame(obj)
_calendar = Cal.calendar(freq=freq) _calendar = Cal.calendar(freq=freq)
data.index = _calendar[data.index.values.astype(np.int)] data.index = _calendar[data.index.values.astype(int)]
data.index.names = ["datetime"] data.index.names = ["datetime"]
if spans is None: if spans is None:
return data return data
else: else:
mask = np.zeros(len(data), dtype=np.bool) mask = np.zeros(len(data), dtype=bool)
for begin, end in spans: for begin, end in spans:
mask |= (data.index >= begin) & (data.index <= end) mask |= (data.index >= begin) & (data.index <= end)
return data[mask] return data[mask]