033.7

trainmodel updated for 3.13

Author: bartzbeielstein

pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "spotpython"
-version = "0.33.6"
+version = "0.33.7"
 authors = [
   { name="T. Bartz-Beielstein", email="tbb@bartzundbartz.de" }
 ]
@@ -45,9 +45,7 @@ dependencies = [
   "plotly",
   "pytest",
   "pytest-mock",
-  "PyQt6",
   "python-markdown-math",
-  "pytorch-lightning>=1.4",
   "river>=0.22.0",
   "scikit-learn",
   "scipy",

src/spotpython/light/trainmodel.py

@@ -129,6 +129,7 @@ def train_model(config: dict, fun_control: dict, timestamp: bool = True) -> floa
         )
     else:
         dm = fun_control["data_module"]
+    dm.setup()  # Manually call setup to prepare the datasets
 
     model = build_model_instance(config, fun_control)
     # TODO: Check if this is necessary or if this is handled by the trainer
@@ -238,7 +239,7 @@ def train_model(config: dict, fun_control: dict, timestamp: bool = True) -> floa
                 gradient_clip_algorithm="norm",
             )
 
-            trainer.fit(model=model, train_dataloaders=train_dl, ckpt_path=None)
+            trainer.fit(model=model, train_dataloaders=train_dl, val_dataloaders=test_dl, ckpt_path=None)
             result = trainer.validate(model=model, dataloaders=test_dl, ckpt_path=None, verbose=verbose)
             result = result[0]
 
@@ -350,10 +351,13 @@ def train_model(config: dict, fun_control: dict, timestamp: bool = True) -> floa
     #   Could also be one of two special keywords "last" and "hpc".
     #   If there is no checkpoint file at the path, an exception is raised.
     try:
-        trainer.fit(model=model, datamodule=dm, ckpt_path=None)
+        trainer.fit(model=model, train_dataloaders=dm.train_dataloader(), val_dataloaders=dm.val_dataloader(), ckpt_path=None)
     except Exception as e:
         print(f"train_model(): trainer.fit failed with exception: {e}")
+        return None
     # Test best model on validation and test set
+    # The validate and test methods expect a datamodule or dataloaders.
+    # Using the datamodule is cleaner.
     verbose = fun_control["verbosity"] > 0
 
     # Validate the model
@@ -455,6 +459,7 @@ def train_model_xai(config: dict, fun_control: dict, timestamp: bool = True) ->
         )
     else:
         dm = fun_control["data_module"]
+    dm.setup()  # Manually call setup to prepare the datasets
 
     model = build_model_instance(config, fun_control)
     # TODO: Check if this is necessary or if this is handled by the trainer
@@ -619,10 +624,13 @@ def train_model_xai(config: dict, fun_control: dict, timestamp: bool = True) ->
     #   Could also be one of two special keywords "last" and "hpc".
     #   If there is no checkpoint file at the path, an exception is raised.
     try:
-        trainer.fit(model=model, datamodule=dm, ckpt_path=None)
+        trainer.fit(model=model, train_dataloaders=dm.train_dataloader(), val_dataloaders=dm.val_dataloader(), ckpt_path=None)
     except Exception as e:
         print(f"train_model(): trainer.fit failed with exception: {e}")
+        return None
     # Test best model on validation and test set
+    # The validate and test methods expect a datamodule or dataloaders.
+    # Using the datamodule is cleaner.
     verbose = fun_control["verbosity"] > 0
 
     # Validate the model