Skip to content

Instantly share code, notes, and snippets.

@hvaara
Created August 27, 2024 16:46
Show Gist options
  • Save hvaara/2b7a48e4ea0a2070bbfd4e914e5c4906 to your computer and use it in GitHub Desktop.
Save hvaara/2b7a48e4ea0a2070bbfd4e914e5c4906 to your computer and use it in GitHub Desktop.
python test/test_modules.py -v -k test_memory_format_nn_BatchNorm2d_
test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float32 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float32) ... ok
test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float64 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_eval_mode_cpu_float64) ... ok
test_memory_format_nn_BatchNorm2d_train_mode_cpu_float32 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_train_mode_cpu_float32) ... ok
test_memory_format_nn_BatchNorm2d_train_mode_cpu_float64 (__main__.TestModuleCPU.test_memory_format_nn_BatchNorm2d_train_mode_cpu_float64) ... ok
test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16) ... FAIL
test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32) ... FAIL
test_memory_format_nn_BatchNorm2d_train_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16) ... FAIL
test_memory_format_nn_BatchNorm2d_train_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32) ... FAIL
======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2918, in wrapper
method(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 447, in instantiated_test
result = test(self, **param_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1515, in wrapper
fn(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 130, in test_wrapper
return test(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 225, in wrapped
return f(*args, **kwargs)
^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 745, in test_memory_format
self.assertEqual(outputs, desired_outputs, rtol=rtol, atol=atol)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3824, in assertEqual
raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!
Mismatched elements: 212 / 216 (98.1%)
Greatest absolute difference: 17.25 at index (0, 1, 1, 3) (up to 1e-05 allowed)
Greatest relative difference: 69.0 at index (1, 2, 2, 5) (up to 0.001 allowed)
To execute this test, run the following from the base repo dir:
python test/test_modules.py TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float16
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2918, in wrapper
method(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 447, in instantiated_test
result = test(self, **param_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1515, in wrapper
fn(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 130, in test_wrapper
return test(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 225, in wrapped
return f(*args, **kwargs)
^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 745, in test_memory_format
self.assertEqual(outputs, desired_outputs, rtol=rtol, atol=atol)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3824, in assertEqual
raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!
Mismatched elements: 212 / 216 (98.1%)
Greatest absolute difference: 16.83740234375 at index (0, 0, 5, 0) (up to 1e-05 allowed)
Greatest relative difference: 178.90695190429688 at index (0, 1, 4, 3) (up to 1e-05 allowed)
To execute this test, run the following from the base repo dir:
python test/test_modules.py TestModuleMPS.test_memory_format_nn_BatchNorm2d_eval_mode_mps_float32
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_train_mode_mps_float16 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2918, in wrapper
method(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 447, in instantiated_test
result = test(self, **param_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1515, in wrapper
fn(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 130, in test_wrapper
return test(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 225, in wrapped
return f(*args, **kwargs)
^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 745, in test_memory_format
self.assertEqual(outputs, desired_outputs, rtol=rtol, atol=atol)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3824, in assertEqual
raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!
Mismatched elements: 216 / 216 (100.0%)
Greatest absolute difference: 3.13671875 at index (0, 0, 2, 3) (up to 1e-05 allowed)
Greatest relative difference: 1662.0 at index (0, 0, 3, 1) (up to 0.001 allowed)
To execute this test, run the following from the base repo dir:
python test/test_modules.py TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float16
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
======================================================================
FAIL: test_memory_format_nn_BatchNorm2d_train_mode_mps_float32 (__main__.TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 2918, in wrapper
method(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_device_type.py", line 447, in instantiated_test
result = test(self, **param_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 1515, in wrapper
fn(*args, **kwargs)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_modules.py", line 130, in test_wrapper
return test(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_cuda.py", line 225, in wrapped
return f(*args, **kwargs)
^^^^^^^^^^^^^^^^^^
File "/Users/hvaara/dev/pytorch/pytorch/test/test_modules.py", line 745, in test_memory_format
self.assertEqual(outputs, desired_outputs, rtol=rtol, atol=atol)
File "/Users/hvaara/dev/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 3824, in assertEqual
raise error_metas.pop()[0].to_error(
AssertionError: Tensor-likes are not close!
Mismatched elements: 216 / 216 (100.0%)
Greatest absolute difference: 3.2349281311035156 at index (0, 0, 5, 0) (up to 1e-05 allowed)
Greatest relative difference: 1475.7142333984375 at index (1, 0, 5, 2) (up to 1e-05 allowed)
To execute this test, run the following from the base repo dir:
python test/test_modules.py TestModuleMPS.test_memory_format_nn_BatchNorm2d_train_mode_mps_float32
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
----------------------------------------------------------------------
Ran 8 tests in 0.202s
FAILED (failures=4)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment