Fix

lighter/callbacks/writer/file.py

@@ -20,15 +20,15 @@ class LighterFileWriter(LighterBaseWriter):
     for a more permanent solution, it can be added to the `self.writers` dictionary.
 
     Args:
-        directory (Union[str, Path]): Directory where the files should be written.
+        path (Union[str, Path]): Directory where the files should be written.
         writer (Union[str, Callable]): Name of the writer function registered in `self.writers` or a custom writer function.
             Available writers: "tensor", "image", "video", "itk_nrrd", "itk_seg_nrrd", "itk_nifti".
             A custom writer function must take two arguments: `path` and `tensor`, and write the tensor to the specified path.
             `tensor` is a single tensor without the batch dimension.
     """
 
-    def __init__(self, directory: Union[str, Path], writer: Union[str, Callable]) -> None:
-        super().__init__(directory, writer)
+    def __init__(self, path: Union[str, Path], writer: Union[str, Callable]) -> None:
+        super().__init__(path, writer)
 
     @property
     def writers(self) -> Dict[str, Callable]:
@@ -49,9 +49,11 @@ def write(self, tensor: torch.Tensor, id: Union[int, str]) -> None:
             tensor (Tensor): Tensor, without the batch dimension, to be written.
             id (Union[int, str]): Identifier, used for file-naming.
         """
+        if not self.path.is_dir():
+            raise RuntimeError(f"LighterFileWriter expects a directory path, got {self.path}")
+
         # Determine the path for the file based on prediction count. The suffix must be added by the writer function.
-        path = self.directory / str(id)
-        path.parent.mkdir(exist_ok=True, parents=True)
+        path = self.path / str(id)
         # Write the tensor to the file.
         self.writer(path, tensor)
 

lighter/utils/dynamic_imports.py

@@ -66,7 +66,8 @@ def import_module_from_path(module_name: str, module_path: str) -> None:
     # Based on https://stackoverflow.com/a/41595552.
 
     if module_name in sys.modules:
-        raise ValueError(f"{module_name} has already been imported as module.")
+        logger.warning(f"{module_name} has already been imported as module.")
+        return
 
     module_path = Path(module_path).resolve() / "__init__.py"
     if not module_path.is_file():

projects/cifar10/experiments/monai_bundle_prototype.yaml

@@ -10,7 +10,7 @@ trainer:
     logger: False
     callbacks:
         - _target_: lighter.callbacks.LighterFileWriter
-          directory: '$f"{@project}/predictions"'
+          path: '$f"{@project}/predictions"'
           writer: tensor
 
 system:

tests/integration/test_cifar.py

@@ -14,13 +14,23 @@
             "fit",
             # Config fiile
             "./projects/cifar10/experiments/monai_bundle_prototype.yaml",
-        )
+        ),
+        (  # Method name
+            "test",
+            # Config fiile
+            "./projects/cifar10/experiments/monai_bundle_prototype.yaml",
+        ),
+        (  # Method name
+            "predict",
+            # Config fiile
+            "./projects/cifar10/experiments/monai_bundle_prototype.yaml",
+        ),
     ],
 )
 @pytest.mark.slow
 def test_trainer_method(method_name: str, config: str):
     """ """
-    kwargs = {"config": [config, test_overrides]}
 
+    kwargs = {"config": [config, test_overrides]}
     func_return = run(method_name, **kwargs)
     assert func_return is None

tests/integration/test_overrides.yaml

@@ -2,4 +2,3 @@ trainer#fast_dev_run: True
 trainer#accelerator: cpu
 system#batch_size: 16
 system#num_workers: 2
-trainer#callbacks: null