Failing XPU Tests Thread

[frost-intel]

Failed tests:

FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_pretrained_pos[True] - UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ModelWeights.Sentinel`. You can also use `weights=ModelWeights.DEFAULT` to get the most up-to-date weights.
FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_pretrained_pos[False] - UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.
FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_equivalent_behavior_weights[True-True] - UserWarning: Using 'weights' as positional parameter(s) is deprecated since 0.13 and may be removed in the future. Please use keyword parameter(s) instead.
FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_equivalent_behavior_weights[True-False] - UserWarning: Using 'weights' as positional parameter(s) is deprecated since 0.13 and may be removed in the future. Please use keyword parameter(s) instead.
FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_equivalent_behavior_weights[False-True] - UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
FAILED test/test_extended_models.py::TestHandleLegacyInterface::test_equivalent_behavior_weights[False-False] - UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
FAILED test/test_functional_tensor.py::test_gaussian_blur[gaussian_blur-sigma3-ksize2-dt3-large-xpu] - AssertionError: (23, 23), (1.7, 1.7)
FAILED test/test_functional_tensor.py::test_gaussian_blur[ScriptFunction-sigma3-ksize2-dt3-large-xpu] - AssertionError: (23, 23), (1.7, 1.7)
FAILED test/test_image.py::test_decode_bad_encoded_data[decode_avif-BMFF parsing failed] - AssertionError: Regex pattern did not match.
    Regex: 'Input tensor must be 1-dimensional'
    Input: "In order to enable the AVIF and HEIC decoding capabilities of torchvision, you need to `pip install torchvision-extra-decoders`. Just install the package, you don't need to update your code. This is only supported on Linux, and this feature is still in BETA stage. Please let us know of any issue: https://github.com/pytorch/vision/issues/new/choose. Note that `torchvision-extra-decoders` is released under the LGPL license. "
FAILED test/test_image.py::test_decode_bad_encoded_data[decode_heic-Invalid input: No 'ftyp' box] - AssertionError: Regex pattern did not match.
    Regex: 'Input tensor must be 1-dimensional'
    Input: "In order to enable the AVIF and HEIC decoding capabilities of torchvision, you need to `pip install torchvision-extra-decoders`. Just install the package, you don't need to update your code. This is only supported on Linux, and this feature is still in BETA stage. Please let us know of any issue: https://github.com/pytorch/vision/issues/new/choose. Note that `torchvision-extra-decoders` is released under the LGPL license. "
FAILED test/test_image.py::test_decode_avif[decode_avif] - RuntimeError: In order to enable the AVIF and HEIC decoding capabilities of torchvision, you need to `pip install torchvision-extra-decoders`. Just install the package, you don't need to update your code. This is only supported on Linux, and this feature is still in BETA stage. Please let us know of any issue: https://github.com/pytorch/vision/issues/new/choose. Note that `torchvision-extra-decoders` is released under the LGPL license.
FAILED test/test_image.py::test_decode_heic[decode_heic] - RuntimeError: In order to enable the AVIF and HEIC decoding capabilities of torchvision, you need to `pip install torchvision-extra-decoders`. Just install the package, you don't need to update your code. This is only supported on Linux, and this feature is still in BETA stage. Please let us know of any issue: https://github.com/pytorch/vision/issues/new/choose. Note that `torchvision-extra-decoders` is released under the LGPL license.
FAILED test/test_models.py::test_classification_model[xpu-inception_v3] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 2 / 50 (4.0%)
    Greatest absolute difference: 26624.0 at index (0, 13) (up to 1e-05 allowed)
    Greatest relative difference: 7.238957095978549e-06 at index (0, 13) (up to 1.3e-06 allowed)
FAILED test/test_models.py::test_classification_model[xpu-resnet101] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 2 / 50 (4.0%)
    Greatest absolute difference: 0.00244140625 at index (0, 26) (up to 1e-05 allowed)
    Greatest relative difference: 1.9300665371702053e-05 at index (0, 22) (up to 1.3e-06 allowed)
FAILED test/test_models.py::test_classification_model[xpu-resnet152] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 4 / 50 (8.0%)
    Greatest absolute difference: 4.25 at index (0, 3) (up to 1e-05 allowed)
    Greatest relative difference: 4.502650426729815e-06 at index (0, 15) (up to 1.3e-06 allowed)
FAILED test/test_models.py::test_classification_model[xpu-wide_resnet101_2] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 4 / 50 (8.0%)
    Greatest absolute difference: 0.000518798828125 at index (0, 24) (up to 1e-05 allowed)
    Greatest relative difference: 5.659311227645958e-06 at index (0, 24) (up to 1.3e-06 allowed)
FAILED test/test_ops.py::TestRoIAlign::test_autocast[rois_dtype1-x_dtype0-True-True-xpu] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 4652 / 5000 (93.0%)
    Greatest absolute difference: 0.0019006729125976562 at index (2, 33, 4, 3) (up to 1e-05 allowed)
    Greatest relative difference: 0.009034737013280392 at index (2, 17, 4, 4) (up to 1e-05 allowed)
FAILED test/test_ops.py::TestRoIAlign::test_autocast[rois_dtype1-x_dtype0-True-False-xpu] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 4679 / 5000 (93.6%)
    Greatest absolute difference: 0.0028815269470214844 at index (1, 15, 4, 2) (up to 1e-05 allowed)
    Greatest relative difference: 0.04907805472612381 at index (1, 12, 2, 2) (up to 1e-05 allowed)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype0-roi_pool] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype0-ps_roi_pool] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype0-roi_align] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype0-ps_roi_align] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype1-roi_pool] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype1-ps_roi_pool] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype1-roi_align] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::test_roi_opcheck[True-xpu-dtype1-ps_roi_align] - torch.testing._internal.optests.generate_tests.OpCheckError: opcheck(op, ...): test_autograd_registration failed with autograd_registration_check: NYI devices other than CPU/CUDA, got {'xpu'} (scroll up for stack trace)
FAILED test/test_ops.py::TestDeformConv::test_autograd_registration__test_autocast[dtype0-0-xpu] - torch.testing._internal.optests.generate_tests.OpCheckError: Test generated by `generate_opcheck_tests`, TestDeformConv.test_autograd_registration__test_autocast, failed on operators ['torchvision::deform_conv2d'].
FAILED test/test_ops.py::TestDeformConv::test_autograd_registration__test_forward[0-True-xpu] - torch.testing._internal.optests.generate_tests.OpCheckError: Test generated by `generate_opcheck_tests`, TestDeformConv.test_autograd_registration__test_forward, failed on operators ['torchvision::deform_conv2d'].
FAILED test/test_transforms_tensor.py::test_linear_transformation[xpu] - AssertionError: Tensor-likes are not equal!
    Mismatched elements: 1307 / 2304 (56.7%)
    Greatest absolute difference: 0.046875 at index (0, 12, 22)
    Greatest relative difference: 3.275850986028672e-07 at index (2, 1, 8)
FAILED test/test_transforms_v2.py::TestGaussianBlur::test_functional_image_correctness[xpu-dtype2-dimensions4-kernel_size4-1.7] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 39 / 728 (5.4%)
    Greatest absolute difference: 1.75 at index (0, 7, 5) (up to 1 allowed)
    Greatest relative difference: 0.0109405517578125 at index (0, 7, 5) (up to 0 allowed)
FAILED test/test_transforms_v2.py::TestConvertBoundingBoxFormat::test_correctness[functional-xpu-dtype1-BoundingBoxFormat.CXCYWHR-BoundingBoxFormat.XYWHR] - AssertionError: Tensor-likes are not equal!
    Mismatched elements: 1 / 5 (20.0%)
    Greatest absolute difference: 9.5367431640625e-07 at index (0, 1)
    Greatest relative difference: 9.649214405271778e-08 at index (0, 1)
FAILED test/test_transforms_v2.py::TestConvertBoundingBoxFormat::test_correctness[functional-xpu-dtype1-BoundingBoxFormat.CXCYWHR-BoundingBoxFormat.XYXYXYXY] - AssertionError: Tensor-likes are not equal!
    Mismatched elements: 4 / 8 (50.0%)
    Greatest absolute difference: 1.9073486328125e-06 at index (0, 7)
    Greatest relative difference: 1.808648562473536e-06 at index (0, 3)
FAILED test/test_transforms_v2.py::TestConvertBoundingBoxFormat::test_correctness[transform-xpu-dtype1-BoundingBoxFormat.CXCYWHR-BoundingBoxFormat.XYWHR] - AssertionError: Tensor-likes are not equal!
    Mismatched elements: 1 / 5 (20.0%)
    Greatest absolute difference: 9.5367431640625e-07 at index (0, 1)
    Greatest relative difference: 9.649214405271778e-08 at index (0, 1)
FAILED test/test_transforms_v2.py::TestConvertBoundingBoxFormat::test_correctness[transform-xpu-dtype1-BoundingBoxFormat.CXCYWHR-BoundingBoxFormat.XYXYXYXY] - AssertionError: Tensor-likes are not equal!
    Mismatched elements: 4 / 8 (50.0%)
    Greatest absolute difference: 1.9073486328125e-06 at index (0, 7)
    Greatest relative difference: 1.808648562473536e-06 at index (0, 3)
FAILED test/test_transforms_v2.py::TestLinearTransform::test_transform[xpu-dtype0-make_image_tensor] - AssertionError: Tensor-likes are not close!
    Mismatched elements: 18 / 561 (3.2%)
    Greatest absolute difference: 0.00079345703125 at index (1, 6, 8) (up to 1e-05 allowed)
    Greatest relative difference: 3.723319969139993e-05 at index (2, 10, 0) (up to 1.3e-06 allowed)

[frost-intel]

PR 8946 will solve the following:

FAILED test/test_ops.py::TestRoiPool::test_autocast[rois_dtype0-x_dtype1-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUHalfType does not equal XPUFloatType (while checking arguments for roi_pool_forward_kernel)
FAILED test/test_ops.py::TestRoiPool::test_autocast[rois_dtype1-x_dtype0-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUFloatType does not equal XPUHalfType (while checking arguments for roi_pool_forward_kernel)
FAILED test/test_ops.py::TestPSRoIPool::test_autocast[rois_dtype0-x_dtype1-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUHalfType does not equal XPUFloatType (while checking arguments for ps_roi_pool_forward_kernel)
FAILED test/test_ops.py::TestPSRoIPool::test_autocast[rois_dtype1-x_dtype0-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUFloatType does not equal XPUHalfType (while checking arguments for ps_roi_pool_forward_kernel)
FAILED test/test_ops.py::TestPSRoIAlign::test_autocast[rois_dtype0-x_dtype1-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUHalfType does not equal XPUFloatType (while checking arguments for ps_roi_align_forward_kernel)
FAILED test/test_ops.py::TestPSRoIAlign::test_autocast[rois_dtype1-x_dtype0-xpu] - RuntimeError: Expected tensor for argument #1 'input' to have the same type as tensor for argument #2 'rois'; but type XPUFloatType does not equal XPUHalfType (while checking arguments for ps_roi_align_forward_kernel)