Lint

Signed-off-by: Krzysztof Lecki <klecki@nvidia.com>

Author: klecki

dali/test/python/nose_utils.py

@@ -12,13 +12,15 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from nose2.tools.decorators import with_setup as _nose2_with_setup, with_teardown as _nose2_with_teardown
+from nose2.tools.decorators import (
+    with_setup as _nose2_with_setup,
+    with_teardown as _nose2_with_teardown,
+)
 from unittest import SkipTest  # noqa: F401
 import unittest
 import re
 import fnmatch
 import functools
-import warnings
 
 
 def with_setup(setup=None, teardown=None):
@@ -31,12 +33,14 @@ def with_setup(setup=None, teardown=None):
         @with_setup(setup_func, teardown_func)
         @with_setup(teardown=teardown_func)
     """
+
     def decorator(func):
         if setup is not None:
             func = _nose2_with_setup(setup)(func)
         if teardown is not None:
             func = _nose2_with_teardown(teardown)(func)
         return func
+
     return decorator
 
 
@@ -51,10 +55,12 @@ def attr(*tags):
     Usage: @attr("pytorch", "slow")
     Filtering: nose2 -A 'pytorch' or nose2 -A '!slow'
     """
+
     def decorator(func):
         for tag in tags:
             setattr(func, tag, True)
         return func
+
     return decorator
 
 

dali/test/python/test_dltensor_operator.py

@@ -183,7 +183,7 @@ def pytorch_red_channel_op(in1, in2):
     return [t.narrow(2, 0, 1).squeeze() for t in in1], [t.narrow(2, 0, 1).squeeze() for t in in2]
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_pytorch():
     setup_pytorch()
     for testcase in [simple_pytorch_op, pytorch_red_channel_op]:
@@ -329,7 +329,7 @@ def cupy_kernel_gray_scale(in1, in2, stream=None):
     return out1, out2
 
 
-@attr('cupy')
+@attr("cupy")
 def test_cupy():
     setup_cupy()
     print(cupy)
@@ -338,7 +338,7 @@ def test_cupy():
     yield from _cupy_flip_with_negative_strides_suite()
 
 
-@attr('cupy')
+@attr("cupy")
 def test_cupy_kernel_gray_scale():
     setup_cupy()
     cupy_case(cupy_kernel_gray_scale, synchronize=False)

dali/test/python/test_fw_iterators_detection.py

@@ -74,7 +74,7 @@ def test_mxnet_pipeline_dynamic_shape():
         assert data is not None
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_pytorch_pipeline_dynamic_shape():
     from nvidia.dali.plugin.pytorch import DALIGenericIterator as PyTorchIterator
 
@@ -87,7 +87,7 @@ def test_pytorch_pipeline_dynamic_shape():
         assert data is not None
 
 
-@attr('paddle')
+@attr("paddle")
 def test_paddle_pipeline_dynamic_shape():
     from nvidia.dali.plugin.paddle import DALIGenericIterator as PaddleIterator
 
@@ -100,7 +100,7 @@ def test_paddle_pipeline_dynamic_shape():
         assert data is not None
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_api_fw_check1_pytorch():
     from nvidia.dali.plugin.pytorch import DALIGenericIterator as PyTorchIterator
 
@@ -120,7 +120,7 @@ def test_api_fw_check1_mxnet():
     )
 
 
-@attr('paddle')
+@attr("paddle")
 def test_api_fw_check1_paddle():
     from nvidia.dali.plugin.paddle import DALIGenericIterator as PaddleIterator
 
@@ -176,14 +176,14 @@ def test_api_fw_check2_mxnet():
     )
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_api_fw_check2_pytorch():
     from nvidia.dali.plugin.pytorch import DALIGenericIterator as PyTorchIterator
 
     yield from test_api_fw_check2(PyTorchIterator, ["data", "bboxes", "label"])
 
 
-@attr('paddle')
+@attr("paddle")
 def test_api_fw_check2_paddle():
     from nvidia.dali.plugin.paddle import DALIGenericIterator as PaddleIterator
 

dali/test/python/test_pytorch_operator.py

@@ -114,7 +114,7 @@ def check_pytorch_operator(device):
             assert numpy.allclose(res2, exp2_t.numpy())
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_pytorch_operator():
     for device in {"cpu", "gpu"}:
         yield check_pytorch_operator, device
@@ -138,7 +138,7 @@ def check_pytorch_operator_batch_processing(device):
             assert numpy.allclose(res2, exp2[i].numpy())
 
 
-@attr('pytorch')
+@attr("pytorch")
 def test_pytorch_operator_batch_processing():
     for device in {"cpu", "gpu"}:
         yield check_pytorch_operator_batch_processing, device