Account for empty validation and test sets

src/metatensor/models/cli/eval.py

@@ -67,6 +67,9 @@ def _add_eval_model_parser(subparser: argparse._SubParsersAction) -> None:
 
 def _eval_targets(model, dataset: Union[_BaseDataset, torch.utils.data.Subset]) -> None:
     """Evaluate an exported model on a dataset and print the RMSEs for each target."""
+    if len(dataset) == 0:
+        logger.info("This dataset is empty")
+        return
     # Attach neighbor lists to the systems:
     requested_neighbor_lists = model.requested_neighbors_lists()
     # working around https://github.com/lab-cosmo/metatensor/issues/521

src/metatensor/models/cli/train.py

@@ -351,6 +351,16 @@ def _train_model_hydra(options: DictConfig) -> None:
             )
             validation_datasets.append(validation_dataset)
 
+    if (
+        sum([len(validation_dataset) for validation_dataset in validation_datasets])
+        == 0
+    ):
+        raise ValueError(
+            "The validation set is empty. Please provide a validation set, "
+            "either by setting a fraction of the training set or by providing it "
+            "explicitly."
+        )
+
     # Save fully expanded config
     OmegaConf.save(config=options, f=Path(output_dir) / "options.yaml")
 

tests/cli/test_train_model.py

@@ -135,6 +135,51 @@ def test_train_multiple_datasets(monkeypatch, tmp_path, options):
     train_model(options)
 
 
+def test_empty_training_set(monkeypatch, tmp_path, options):
+    """Test that an error is raised if no training set is provided."""
+    monkeypatch.chdir(tmp_path)
+
+    shutil.copy(DATASET_PATH, "qm9_reduced_100.xyz")
+
+    options["validation_set"] = 0.6
+    options["test_set"] = 0.4
+
+    with pytest.raises(
+        ValueError, match="Fraction of the train set is smaller or equal to 0!"
+    ):
+        train_model(options)
+
+
+def test_empty_validation_set(monkeypatch, tmp_path, options):
+    """Test that an error is raised if no validation set is provided."""
+    monkeypatch.chdir(tmp_path)
+
+    shutil.copy(DATASET_PATH, "qm9_reduced_100.xyz")
+
+    options["validation_set"] = 0.0
+    options["test_set"] = 0.4
+
+    with pytest.raises(ValueError, match="The validation set is empty. Please provide"):
+        train_model(options)
+
+
+def test_empty_test_set(monkeypatch, tmp_path, options):
+    """Test that no error is raised if no test set is provided."""
+    monkeypatch.chdir(tmp_path)
+
+    shutil.copy(DATASET_PATH, "qm9_reduced_100.xyz")
+
+    options["validation_set"] = 0.4
+    options["test_set"] = 0.0
+
+    train_model(options)
+
+    # check if the logging is correct
+    with open(glob.glob("outputs/*/*/train.log")[0]) as f:
+        log = f.read()
+    assert "This dataset is empty" in log
+
+
 @pytest.mark.parametrize(
     "test_set_file, validation_set_file", [(True, False), (False, True)]
 )