(1) Fix /0 bug in double_ensemble, (2) remove _default_uri for R/expm, (3) support model size in pytorch models

qlib/config.py

@@ -105,7 +105,7 @@ def set_conf_from_C(self, config_c):
     "redis_port": 6379,
     "redis_task_db": 1,
     # This value can be reset via qlib.init
-    "logging_level": "INFO",
+    "logging_level": logging.INFO,
     # Global configuration of qlib log
     # logging_level can control the logging level more finely
     "logging_config": {
@@ -124,12 +124,12 @@ def set_conf_from_C(self, config_c):
         "handlers": {
             "console": {
                 "class": "logging.StreamHandler",
-                "level": "DEBUG",
+                "level": logging.DEBUG,
                 "formatter": "logger_format",
                 "filters": ["field_not_found"],
             }
         },
-        "loggers": {"qlib": {"level": "DEBUG", "handlers": ["console"]}},
+        "loggers": {"qlib": {"level": logging.DEBUG, "handlers": ["console"]}},
     },
     # Defatult config for experiment manager
     "exp_manager": {
@@ -185,7 +185,7 @@ def set_conf_from_C(self, config_c):
         # The nfs should be auto-mounted by qlib on other
         # serversS(such as PAI) [auto_mount:True]
         "timeout": 100,
-        "logging_level": "INFO",
+        "logging_level": logging.INFO,
         "region": REG_CN,
         ## Custom Operator
         "custom_ops": [],

qlib/contrib/model/double_ensemble.py

@@ -184,7 +184,7 @@ def feature_selection(self, df_train, loss_values):
                     / M
                 )
             loss_feat = self.get_loss(y_train.values.squeeze(), pred.values)
-            g.loc[i_f, "g_value"] = np.mean(loss_feat - loss_values) / np.std(loss_feat - loss_values)
+            g.loc[i_f, "g_value"] = np.mean(loss_feat - loss_values) / (np.std(loss_feat - loss_values) + 1e-7)
             x_train_tmp.loc[:, feat] = x_train.loc[:, feat].copy()
 
         # one column in train features is all-nan # if g['g_value'].isna().any()

qlib/contrib/model/pytorch_alstm.py

@@ -23,6 +23,7 @@
 import torch.nn as nn
 import torch.optim as optim
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -39,8 +40,8 @@ class ALSTM(Model):
         the evaluate metric used in early stop
     optimizer : str
         optimizer name
-    GPU : str
-        the GPU ID(s) used for training
+    GPU : int
+        the GPU ID used for training
     """
 
     def __init__(
@@ -76,7 +77,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
 
@@ -123,6 +124,9 @@ def __init__(
             num_layers=self.num_layers,
             dropout=self.dropout,
         )
+        self.logger.info("model:\n{:}".format(self.ALSTM_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.ALSTM_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.ALSTM_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -214,7 +218,6 @@ def fit(
         self,
         dataset: DatasetH,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
 

qlib/contrib/model/pytorch_alstm_ts.py

@@ -24,6 +24,7 @@
 import torch.optim as optim
 from torch.utils.data import DataLoader
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH, TSDatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -40,8 +41,8 @@ class ALSTM(Model):
         the evaluate metric used in early stop
     optimizer : str
         optimizer name
-    GPU : str
-        the GPU ID(s) used for training
+    GPU : int
+        the GPU ID used for training
     """
 
     def __init__(
@@ -78,7 +79,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.n_jobs = n_jobs
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
@@ -127,7 +128,10 @@ def __init__(
             hidden_size=self.hidden_size,
             num_layers=self.num_layers,
             dropout=self.dropout,
-        ).to(self.device)
+        )
+        self.logger.info("model:\n{:}".format(self.ALSTM_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.ALSTM_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.ALSTM_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -201,7 +205,6 @@ def fit(
         self,
         dataset,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
         dl_train = dataset.prepare("train", col_set=["feature", "label"], data_key=DataHandlerLP.DK_L)

qlib/contrib/model/pytorch_gats.py

@@ -22,6 +22,7 @@
 import torch.nn as nn
 import torch.optim as optim
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -42,8 +43,8 @@ class GATs(Model):
         the evaluate metric used in early stop
     optimizer : str
         optimizer name
-    GPU : str
-        the GPU ID(s) used for training
+    GPU : int
+        the GPU ID used for training
     """
 
     def __init__(
@@ -83,7 +84,7 @@ def __init__(
         self.base_model = base_model
         self.with_pretrain = with_pretrain
         self.model_path = model_path
-        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() else "cpu")
+        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
 
@@ -135,6 +136,9 @@ def __init__(
             dropout=self.dropout,
             base_model=self.base_model,
         )
+        self.logger.info("model:\n{:}".format(self.GAT_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.GAT_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.GAT_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -232,7 +236,6 @@ def fit(
         self,
         dataset: DatasetH,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
 

qlib/contrib/model/pytorch_gats_ts.py

@@ -24,6 +24,7 @@
 from torch.utils.data import DataLoader
 from torch.utils.data import Sampler
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -62,8 +63,8 @@ class GATs(Model):
         the evaluate metric used in early stop
     optimizer : str
         optimizer name
-    GPU : str
-        the GPU ID(s) used for training
+    GPU : int
+        the GPU ID used for training
     """
 
     def __init__(
@@ -104,7 +105,7 @@ def __init__(
         self.base_model = base_model
         self.with_pretrain = with_pretrain
         self.model_path = model_path
-        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() else "cpu")
+        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.n_jobs = n_jobs
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
@@ -157,6 +158,9 @@ def __init__(
             dropout=self.dropout,
             base_model=self.base_model,
         )
+        self.logger.info("model:\n{:}".format(self.GAT_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.GAT_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.GAT_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -245,7 +249,6 @@ def fit(
         self,
         dataset,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
 

qlib/contrib/model/pytorch_gru.py

@@ -23,6 +23,7 @@
 import torch.nn as nn
 import torch.optim as optim
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -76,7 +77,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
 
@@ -123,6 +124,9 @@ def __init__(
             num_layers=self.num_layers,
             dropout=self.dropout,
         )
+        self.logger.info("model:\n{:}".format(self.gru_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.gru_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.gru_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -214,7 +218,6 @@ def fit(
         self,
         dataset: DatasetH,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
 

qlib/contrib/model/pytorch_gru_ts.py

@@ -24,6 +24,7 @@
 import torch.optim as optim
 from torch.utils.data import DataLoader
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH, TSDatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -78,7 +79,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.n_jobs = n_jobs
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
@@ -127,7 +128,10 @@ def __init__(
             hidden_size=self.hidden_size,
             num_layers=self.num_layers,
             dropout=self.dropout,
-        ).to(self.device)
+        )
+        self.logger.info("model:\n{:}".format(self.gru_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.gru_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.GRU_model.parameters(), lr=self.lr)
         elif optimizer.lower() == "gd":
@@ -201,7 +205,6 @@ def fit(
         self,
         dataset,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
         dl_train = dataset.prepare("train", col_set=["feature", "label"], data_key=DataHandlerLP.DK_L)

qlib/contrib/model/pytorch_lstm.py

@@ -76,7 +76,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
 
@@ -214,7 +214,6 @@ def fit(
         self,
         dataset: DatasetH,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
 

qlib/contrib/model/pytorch_lstm_ts.py

@@ -78,7 +78,7 @@ def __init__(
         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.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.n_jobs = n_jobs
         self.use_gpu = torch.cuda.is_available()
         self.seed = seed
@@ -201,7 +201,6 @@ def fit(
         self,
         dataset,
         evals_result=dict(),
-        verbose=True,
         save_path=None,
     ):
         dl_train = dataset.prepare("train", col_set=["feature", "label"], data_key=DataHandlerLP.DK_L)

qlib/contrib/model/pytorch_nn.py

@@ -15,6 +15,7 @@
 import torch.nn as nn
 import torch.optim as optim
 
+from .pytorch_utils import count_parameters
 from ...model.base import Model
 from ...data.dataset import DatasetH
 from ...data.dataset.handler import DataHandlerLP
@@ -42,8 +43,8 @@ class DNNModelPytorch(Model):
         learning rate decay steps
     optimizer : str
         optimizer name
-    GPU : str
-        the GPU ID(s) used for training
+    GPU : int
+        the GPU ID used for training
     """
 
     def __init__(
@@ -80,7 +81,7 @@ def __init__(
         self.lr_decay_steps = lr_decay_steps
         self.optimizer = optimizer.lower()
         self.loss_type = loss
-        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() else "cpu")
+        self.device = torch.device("cuda:%d" % (GPU) if torch.cuda.is_available() and GPU >= 0 else "cpu")
         self.use_GPU = torch.cuda.is_available()
         self.seed = seed
         self.weight_decay = weight_decay
@@ -129,6 +130,9 @@ def __init__(
         self._scorer = mean_squared_error if loss == "mse" else roc_auc_score
 
         self.dnn_model = Net(input_dim, output_dim, layers, loss=self.loss_type)
+        self.logger.info("model:\n{:}".format(self.dnn_model))
+        self.logger.info("model size: {:.4f} MB".format(count_parameters(self.dnn_model)))
+
         if optimizer.lower() == "adam":
             self.train_optimizer = optim.Adam(self.dnn_model.parameters(), lr=self.lr, weight_decay=self.weight_decay)
         elif optimizer.lower() == "gd":