Merge pull request #123 from k-dominik/fix-win-compat-long

Use 'int64' instead of non-portable 'long'

.github/workflows/conda-build.yml

@@ -9,21 +9,20 @@ on:
     branches: [master]
 
 jobs:
-  build-linux:
-    runs-on: ubuntu-latest
-
+  build:
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [macos-latest, windows-latest, ubuntu-latest]
+    runs-on: ${{ matrix.os }}
     steps:
     - uses: actions/checkout@v2
-    - name: Set pytorch-3dunet version
-      run: echo "RELEASE_VERSION=$(date +%Y%m%d%H%M%S)" >> $GITHUB_ENV
-    - name: Print pytorch-3dunet version
-      run: echo $RELEASE_VERSION
     - uses: conda-incubator/setup-miniconda@v2
       with:
         auto-activate-base: true
         activate-environment: ""
-        channels: local,conda-forge,defaults
-        channel-priority: false
+        channel-priority: flexible
+        miniforge-variant: Miniforge3
     - shell: bash -l {0}
       run: conda info --envs
     - name: Build pytorch-3dunet
@@ -33,15 +32,15 @@ jobs:
         conda build -c pytorch -c nvidia -c conda-forge conda-recipe
     - name: Create pytorch3dunet env
       run: |
-        conda create -n pytorch3dunet -c pytorch -c nvidia -c conda-forge pytorch-3dunet pytest
+        conda create -n pytorch3dunet -c local -c pytorch -c nvidia -c conda-forge pytorch-3dunet pytest
     - name: Run pytest
       shell: bash -l {0}
       run: |
         conda activate pytorch3dunet
         pytest
         conda deactivate
     - name: Deploy on conda
-      if: ${{ startsWith( github.ref, 'refs/tags/') && success() }}
+      if: ${{ matrix.os == 'ubuntu-latest' && startsWith( github.ref, 'refs/tags/') && success() }}
       env:
         ANACONDA_SECRET: ${{ secrets.ANACONDA_TOKEN }}
       shell: bash -l {0}

conda-recipe/meta.yaml

@@ -1,6 +1,8 @@
+{% set setup_py_data = load_setup_py_data() %}
+
 package:
   name: pytorch-3dunet
-  version: {{ RELEASE_VERSION }}
+  version: {{ setup_py_data['version'] }}
 
 source:
   path: ..

resources/3DUnet_multiclass/train_config.yaml

@@ -129,7 +129,7 @@ loaders:
           # convert target volume to binary mask
         - name: ToTensor
           expand_dims: false
-          dtype: long
+          dtype: int64
 
   # configuration of the val loader
   val:
@@ -158,4 +158,4 @@ loaders:
       label:
         - name: ToTensor
           expand_dims: false
-          dtype: long
+          dtype: int64

tests/resources/config_train.yml

@@ -54,8 +54,8 @@ loaders:
         - name: ToTensor
           # do not expand dims for cross-entropy loss
           expand_dims: false
-          # cross-entropy loss requires target to be of type 'long'
-          dtype: 'long'
+          # cross-entropy loss requires target to be of type 'int64'
+          dtype: 'int64'
       weight:
         - name: ToTensor
           expand_dims: false
@@ -76,7 +76,7 @@ loaders:
       label:
         - name: ToTensor
           expand_dims: false
-          dtype: 'long'
+          dtype: 'int64'
       weight:
         - name: ToTensor
           expand_dims: false

tests/resources/config_train_2d.yml

@@ -76,8 +76,8 @@ loaders:
         - name: ToTensor
           # do not expand dims for cross-entropy loss
           expand_dims: false
-          # cross-entropy loss requires target to be of type 'long'
-          dtype: 'long'
+          # cross-entropy loss requires target to be of type 'int64'
+          dtype: 'int64'
       weight:
         - name: ToTensor
           expand_dims: false
@@ -99,7 +99,7 @@ loaders:
       label:
         - name: ToTensor
           expand_dims: false
-          dtype: 'long'
+          dtype: 'int64'
       weight:
         - name: ToTensor
           expand_dims: false

tests/test_dataset.py

@@ -1,4 +1,5 @@
 import os
+from pathlib import Path
 from tempfile import NamedTemporaryFile
 
 import h5py
@@ -75,20 +76,19 @@ def test_lazy_hdf5_dataset(self, transformer_config):
             assert np.all(visit_raw)
             assert np.all(visit_label)
 
-    def test_augmentation(self, transformer_config):
+    def test_augmentation(self, transformer_config, tmpdir):
         raw = np.random.rand(32, 96, 96)
         # assign raw to label's channels for ease of comparison
         label = np.stack([raw for _ in range(3)])
         # create temporary h5 file
-        tmp_file = NamedTemporaryFile()
-        tmp_path = tmp_file.name
-        with h5py.File(tmp_path, 'w') as f:
+        tmp_file = tmpdir / "test.h5"
+        with h5py.File(tmp_file, 'w') as f:
             f.create_dataset('raw', data=raw)
             f.create_dataset('label', data=label)
 
         # set phase='train' in order to execute the train transformers
         phase = 'train'
-        dataset = StandardHDF5Dataset(tmp_path, phase=phase,
+        dataset = StandardHDF5Dataset(tmp_file, phase=phase,
                                       slice_builder_config=_slice_builder_conf((16, 64, 64), (8, 32, 32)),
                                       transformer_config=transformer_config[phase]['transformer'])
 
@@ -120,19 +120,19 @@ def test_traverse_file_paths(self, tmpdir):
 
         assert expected_files == actual_files
 
-    def test_halo(self):
+    def test_halo(self, tmpdir: Path):
         halo_shape = (1, 2, 3)
 
         # create temporary h5 file
         raw = np.random.rand(32, 96, 96)
-        tmp_file = NamedTemporaryFile()
-        tmp_path = tmp_file.name
-        with h5py.File(tmp_path, 'w') as f:
+        tmp_file = tmpdir / "test.h5"
+
+        with h5py.File(tmp_file, 'w') as f:
             f.create_dataset('raw', data=raw)
 
         # halo only implemented with test phase
         phase = 'test'
-        dataset = StandardHDF5Dataset(tmp_path, phase=phase,
+        dataset = StandardHDF5Dataset(tmp_file, phase=phase,
                                       slice_builder_config=_slice_builder_conf((16, 64, 64), (8, 32, 32), halo_shape),
                                       transformer_config=_transformer_test_conf())
         data_loader = DataLoader(dataset, batch_size=1, num_workers=4, shuffle=True,

tests/test_transforms.py

@@ -5,7 +5,7 @@
 
 
 class TestTransforms:
-    config = {'dtype': 'long'}
+    config = {'dtype': 'int64'}
 
     def test_random_label_to_boundary(self):
         size = 20
@@ -62,7 +62,7 @@ def test_relabel(self):
     def test_BaseTransformer(self):
         config = {
             'raw': [{'name': 'Standardize'}, {'name': 'ToTensor', 'expand_dims': True}],
-            'label': [{'name': 'ToTensor', 'expand_dims': False, 'dtype': 'long'}],
+            'label': [{'name': 'ToTensor', 'expand_dims': False, 'dtype': 'int64'}],
             'weight': [{'name': 'ToTensor', 'expand_dims': False}]
         }
         base_config = {'mean': 0, 'std': 1}
@@ -73,7 +73,7 @@ def test_BaseTransformer(self):
         assert raw_transforms[1].expand_dims
         label_transforms = transformer.label_transform().transforms
         assert not label_transforms[0].expand_dims
-        assert label_transforms[0].dtype == 'long'
+        assert label_transforms[0].dtype == 'int64'
         weight_transforms = transformer.weight_transform().transforms
         assert not weight_transforms[0].expand_dims
 
@@ -89,7 +89,7 @@ def test_StandardTransformer(self):
             'label': [
                 {'name': 'RandomFlip'},
                 {'name': 'RandomRotate90'},
-                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'long'}
+                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'int64'}
             ]
         }
         base_config = {'mean': 0, 'std': 1}
@@ -116,7 +116,7 @@ def test_AnisotropicRotationTransformer(self):
                 {'name': 'RandomFlip'},
                 {'name': 'RandomRotate90'},
                 {'name': 'RandomRotate', 'angle_spectrum': 17, 'axes': [[2, 1]]},
-                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'long'}
+                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'int64'}
             ]
         }
         base_config = {'mean': 0, 'std': 1}
@@ -145,7 +145,7 @@ def test_LabelToBoundaryTransformer(self):
                 {'name': 'RandomRotate90'},
                 {'name': 'RandomRotate', 'angle_spectrum': 17, 'axes': [[2, 1]], 'mode': 'reflect'},
                 {'name': 'LabelToAffinities', 'offsets': [2, 4, 6, 8]},
-                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'long'}
+                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'int64'}
             ]
         }
         base_config = {'mean': 0, 'std': 1}
@@ -179,7 +179,7 @@ def test_RandomLabelToBoundaryTransformer(self):
                 {'name': 'RandomRotate90'},
                 {'name': 'RandomRotate', 'angle_spectrum': 17, 'axes': [[2, 1]], 'mode': 'reflect'},
                 {'name': 'RandomLabelToAffinities', 'max_offset': 4},
-                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'long'}
+                {'name': 'ToTensor', 'expand_dims': False, 'dtype': 'int64'}
             ]
         }
         base_config = {'mean': 0, 'std': 1}