6136 6146 update the default writer flag (#6147)

Fixes #6136 
fixes #6146 


### Types of changes
<!--- Put an `x` in all the boxes that apply, and remove the not
applicable items -->
- [x] Non-breaking change (fix or new feature that would not break
existing functionality).
- [ ] Breaking change (fix or new feature that would cause existing
functionality to change).
- [x] New tests added to cover the changes.
- [ ] Integration tests passed locally by running `./runtests.sh -f -u
--net --coverage`.
- [x] Quick tests passed locally by running `./runtests.sh --quick
--unittests --disttests`.
- [x] In-line docstrings updated.
- [ ] Documentation updated, tested `make html` command in the `docs/`
folder.

---------

Signed-off-by: Wenqi Li <[email protected]>

Author: wyli

monai/data/image_writer.py

@@ -373,13 +373,14 @@ class ITKWriter(ImageWriter):
     output_dtype: DtypeLike = None
     channel_dim: int | None
 
-    def __init__(self, output_dtype: DtypeLike = np.float32, affine_lps_to_ras: bool = True, **kwargs):
+    def __init__(self, output_dtype: DtypeLike = np.float32, affine_lps_to_ras: bool | None = True, **kwargs):
         """
         Args:
             output_dtype: output data type.
             affine_lps_to_ras: whether to convert the affine matrix from "LPS" to "RAS". Defaults to ``True``.
                 Set to ``True`` to be consistent with ``NibabelWriter``,
                 otherwise the affine matrix is assumed already in the ITK convention.
+                Set to ``None`` to use ``data_array.meta[MetaKeys.SPACE]`` to determine the flag.
             kwargs: keyword arguments passed to ``ImageWriter``.
 
         The constructor will create ``self.output_dtype`` internally.
@@ -406,17 +407,20 @@ def set_data_array(
             kwargs: keyword arguments passed to ``self.convert_to_channel_last``,
                 currently support ``spatial_ndim`` and ``contiguous``, defauting to ``3`` and ``False`` respectively.
         """
-        _r = len(data_array.shape)
+        n_chns = data_array.shape[channel_dim] if channel_dim is not None else 0
         self.data_obj = self.convert_to_channel_last(
             data=data_array,
             channel_dim=channel_dim,
             squeeze_end_dims=squeeze_end_dims,
             spatial_ndim=kwargs.pop("spatial_ndim", 3),
             contiguous=kwargs.pop("contiguous", True),
         )
-        self.channel_dim = (
-            channel_dim if self.data_obj is not None and len(self.data_obj.shape) >= _r else None
-        )  # channel dim is at the end
+        self.channel_dim = -1  # in most cases, the data is set to channel last
+        if squeeze_end_dims and n_chns <= 1:  # num_channel==1 squeezed
+            self.channel_dim = None
+        if not squeeze_end_dims and n_chns < 1:  # originally no channel and convert_to_channel_last added a channel
+            self.channel_dim = None
+            self.data_obj = self.data_obj[..., 0]
 
     def set_metadata(self, meta_dict: Mapping | None = None, resample: bool = True, **options):
         """
@@ -478,7 +482,7 @@ def create_backend_obj(
         channel_dim: int | None = 0,
         affine: NdarrayOrTensor | None = None,
         dtype: DtypeLike = np.float32,
-        affine_lps_to_ras: bool = True,
+        affine_lps_to_ras: bool | None = True,
         **kwargs,
     ):
         """
@@ -492,14 +496,18 @@ def create_backend_obj(
             affine_lps_to_ras: whether to convert the affine matrix from "LPS" to "RAS". Defaults to ``True``.
                 Set to ``True`` to be consistent with ``NibabelWriter``,
                 otherwise the affine matrix is assumed already in the ITK convention.
+                Set to ``None`` to use ``data_array.meta[MetaKeys.SPACE]`` to determine the flag.
             kwargs: keyword arguments. Current `itk.GetImageFromArray` will read ``ttype`` from this dictionary.
 
         see also:
 
             - https://github.com/InsightSoftwareConsortium/ITK/blob/v5.2.1/Wrapping/Generators/Python/itk/support/extras.py#L389
+
         """
-        if isinstance(data_array, MetaTensor) and data_array.meta.get(MetaKeys.SPACE, SpaceKeys.LPS) != SpaceKeys.LPS:
-            affine_lps_to_ras = False  # do the converting from LPS to RAS only if the space type is currently LPS.
+        if isinstance(data_array, MetaTensor) and affine_lps_to_ras is None:
+            affine_lps_to_ras = (
+                data_array.meta.get(MetaKeys.SPACE, SpaceKeys.LPS) != SpaceKeys.LPS
+            )  # do the converting from LPS to RAS only if the space type is currently LPS.
         data_array = super().create_backend_obj(data_array)
         _is_vec = channel_dim is not None
         if _is_vec:

monai/networks/layers/filtering.py

@@ -32,13 +32,10 @@ class BilateralFilter(torch.autograd.Function):
 
     Args:
         input: input tensor.
-
         spatial_sigma: the standard deviation of the spatial blur. Higher values can
             hurt performance when not using the approximate method (see fast approx).
-
         color_sigma: the standard deviation of the color blur. Lower values preserve
             edges better whilst higher values tend to a simple gaussian spatial blur.
-
         fast approx: This flag chooses between two implementations. The approximate method may
             produce artifacts in some scenarios whereas the exact solution may be intolerably
             slow for high spatial standard deviations.
@@ -76,9 +73,7 @@ class PHLFilter(torch.autograd.Function):
 
     Args:
         input: input tensor to be filtered.
-
         features: feature tensor used to filter the input.
-
         sigmas: the standard deviations of each feature in the filter.
 
     Returns:
@@ -114,13 +109,9 @@ class TrainableBilateralFilterFunction(torch.autograd.Function):
 
     Args:
         input: input tensor to be filtered.
-
         sigma x: trainable standard deviation of the spatial filter kernel in x direction.
-
         sigma y: trainable standard deviation of the spatial filter kernel in y direction.
-
         sigma z: trainable standard deviation of the spatial filter kernel in z direction.
-
         color sigma: trainable standard deviation of the intensity range kernel. This filter
             parameter determines the degree of edge preservation.
 
@@ -200,11 +191,9 @@ class TrainableBilateralFilter(torch.nn.Module):
 
     Args:
         input: input tensor to be filtered.
-
         spatial_sigma: tuple (sigma_x, sigma_y, sigma_z) initializing the trainable standard
             deviations of the spatial filter kernels. Tuple length must equal the number of
             spatial input dimensions.
-
         color_sigma: trainable standard deviation of the intensity range kernel. This filter
             parameter determines the degree of edge preservation.
 
@@ -280,15 +269,10 @@ class TrainableJointBilateralFilterFunction(torch.autograd.Function):
 
     Args:
         input: input tensor to be filtered.
-
         guide: guidance image tensor to be used during filtering.
-
         sigma x: trainable standard deviation of the spatial filter kernel in x direction.
-
         sigma y: trainable standard deviation of the spatial filter kernel in y direction.
-
         sigma z: trainable standard deviation of the spatial filter kernel in z direction.
-
         color sigma: trainable standard deviation of the intensity range kernel. This filter
             parameter determines the degree of edge preservation.
 
@@ -373,13 +357,10 @@ class TrainableJointBilateralFilter(torch.nn.Module):
 
     Args:
         input: input tensor to be filtered.
-
         guide: guidance image tensor to be used during filtering.
-
         spatial_sigma: tuple (sigma_x, sigma_y, sigma_z) initializing the trainable standard
             deviations of the spatial filter kernels. Tuple length must equal the number of
             spatial input dimensions.
-
         color_sigma: trainable standard deviation of the intensity range kernel. This filter
             parameter determines the degree of edge preservation.
 

monai/transforms/io/array.py

@@ -40,6 +40,7 @@
     PydicomReader,
 )
 from monai.data.meta_tensor import MetaTensor
+from monai.data.utils import is_no_channel
 from monai.transforms.transform import Transform
 from monai.transforms.utility.array import EnsureChannelFirst
 from monai.utils import GridSamplePadMode
@@ -440,6 +441,7 @@ def set_options(self, init_kwargs=None, data_kwargs=None, meta_kwargs=None, writ
             self.meta_kwargs.update(meta_kwargs)
         if write_kwargs is not None:
             self.write_kwargs.update(write_kwargs)
+        return self
 
     def __call__(self, img: torch.Tensor | np.ndarray, meta_data: dict | None = None):
         """
@@ -450,8 +452,15 @@ def __call__(self, img: torch.Tensor | np.ndarray, meta_data: dict | None = None
         meta_data = img.meta if isinstance(img, MetaTensor) else meta_data
         kw = self.fname_formatter(meta_data, self)
         filename = self.folder_layout.filename(**kw)
-        if meta_data and len(ensure_tuple(meta_data.get("spatial_shape", ()))) == len(img.shape):
-            self.data_kwargs["channel_dim"] = None
+        if meta_data:
+            meta_spatial_shape = ensure_tuple(meta_data.get("spatial_shape", ()))
+            if len(meta_spatial_shape) >= len(img.shape):
+                self.data_kwargs["channel_dim"] = None
+            elif is_no_channel(self.data_kwargs.get("channel_dim")):
+                warnings.warn(
+                    f"data shape {img.shape} (with spatial shape {meta_spatial_shape}) "
+                    f"but SaveImage `channel_dim` is set to {self.data_kwargs.get('channel_dim')} no channel."
+                )
 
         err = []
         for writer_cls in self.writers:

monai/transforms/io/dictionary.py

@@ -296,6 +296,7 @@ def __init__(
 
     def set_options(self, init_kwargs=None, data_kwargs=None, meta_kwargs=None, write_kwargs=None):
         self.saver.set_options(init_kwargs, data_kwargs, meta_kwargs, write_kwargs)
+        return self
 
     def __call__(self, data):
         d = dict(data)

tests/test_auto3dseg_ensemble.py

@@ -70,7 +70,7 @@
 
 
 @skip_if_quick
-@SkipIfBeforePyTorchVersion((1, 10, 0))
+@SkipIfBeforePyTorchVersion((1, 13, 0))
 @unittest.skipIf(not has_tb, "no tensorboard summary writer")
 class TestEnsembleBuilder(unittest.TestCase):
     def setUp(self) -> None:

tests/test_image_rw.py

@@ -167,7 +167,7 @@ def nrrd_rw(self, test_data, reader, writer, dtype, resample=True):
             filepath = f"testfile_{ndim}d"
             saver = SaveImage(
                 output_dir=self.test_dir, output_ext=output_ext, resample=resample, separate_folder=False, writer=writer
-            )
+            ).set_options(init_kwargs={"affine_lps_to_ras": True})
             test_data = MetaTensor(
                 p(test_data), meta={"filename_or_obj": f"{filepath}{output_ext}", "spatial_shape": test_data.shape}
             )

tests/test_integration_autorunner.py

@@ -69,7 +69,7 @@
 
 
 @skip_if_quick
-@SkipIfBeforePyTorchVersion((1, 9, 1))
+@SkipIfBeforePyTorchVersion((1, 13, 0))
 @unittest.skipIf(not has_tb, "no tensorboard summary writer")
 class TestAutoRunner(unittest.TestCase):
     def setUp(self) -> None:

tests/test_integration_gpu_customization.py

@@ -69,7 +69,7 @@
 
 
 @skip_if_quick
-@SkipIfBeforePyTorchVersion((1, 9, 1))
+@SkipIfBeforePyTorchVersion((1, 13, 0))
 @unittest.skipIf(not has_tb, "no tensorboard summary writer")
 class TestEnsembleGpuCustomization(unittest.TestCase):
     def setUp(self) -> None:

tests/test_itk_writer.py

@@ -52,6 +52,17 @@ def test_rgb(self):
             np.testing.assert_allclose(output.shape, (5, 5, 3))
             np.testing.assert_allclose(output[1, 1], (5, 5, 4))
 
+    def test_no_channel(self):
+        with tempfile.TemporaryDirectory() as tempdir:
+            fname = os.path.join(tempdir, "testing.nii.gz")
+            writer = ITKWriter(output_dtype=np.uint8)
+            writer.set_data_array(np.arange(48).reshape(3, 4, 4), channel_dim=None)
+            writer.write(fname)
+
+            output = np.asarray(itk.imread(fname))
+            np.testing.assert_allclose(output.shape, (4, 4, 3))
+            np.testing.assert_allclose(output[1, 1], (5, 21, 37))
+
 
 if __name__ == "__main__":
     unittest.main()

tests/testing_data/integration_answers.py

@@ -14,6 +14,62 @@
 import numpy as np
 
 EXPECTED_ANSWERS = [
+    {  # test answers for PyTorch 2.0
+        "integration_segmentation_3d": {
+            "losses": [
+                0.5430086106061935,
+                0.47010003924369814,
+                0.4453376233577728,
+                0.451901963353157,
+                0.4398456811904907,
+                0.43450237810611725,
+            ],
+            "best_metric": 0.9329540133476257,
+            "infer_metric": 0.9330471754074097,
+            "output_sums": [
+                0.14212507078546172,
+                0.15199039602949577,
+                0.15133471939291526,
+                0.13967984811021827,
+                0.18831614355832332,
+                0.1694076821827231,
+                0.14663931509271658,
+                0.16788710637623733,
+                0.1569452710008219,
+                0.17907130698392254,
+                0.16244092698688475,
+                0.1679350345855819,
+                0.14437674754879065,
+                0.11355098478396568,
+                0.161660275855964,
+                0.20082478187698194,
+                0.17575491677668853,
+                0.0974593860605401,
+                0.19366775441539907,
+                0.20293016863409002,
+                0.19610441127101647,
+                0.20812173772459808,
+                0.16184212006067655,
+                0.13185211452732482,
+                0.14824716961304257,
+                0.14229818359602905,
+                0.23141282114085215,
+                0.1609268635938338,
+                0.14825300029123678,
+                0.10286266811772046,
+                0.11873484714087054,
+                0.1296615212510262,
+                0.11386621034856693,
+                0.15203351148564773,
+                0.16300823766585265,
+                0.1936726544485426,
+                0.2227251185536394,
+                0.18067789917505797,
+                0.19005874127683337,
+                0.07462121515702229,
+            ],
+        }
+    },
     {  # test answers for PyTorch 1.12.1
         "integration_classification_2d": {
             "losses": [0.776835828070428, 0.1615355300011149, 0.07492854832938523, 0.04591309238865877],