import numpy as np
import tensorflow as tf
x = tf.placeholder(tf.float32, shape=[1, 3, None, None])
xpad = tf.pad(x, [[0, 0], [0, 0], [2, 3], [2, 3]])
W = tf.random_normal([7, 7, 3, 64])
out = tf.nn.conv2d(xpad, W, [1, 1, 2, 2], padding="VALID", data_format='NCHW')
sess = tf.Session()
with sess.as_default():
sess.run(out, feed_dict=
{x: np.random.rand(1, 3, 600, 800)})This simple code snippet is legal and runs well on a GPU. However, when run on a CPU with MKL build it throws:
2019-01-01 04:17:00.402605: W tensorflow/core/framework/op_kernel.cc:1412] OP_REQUIRES failed at mkl_conv_ops.cc:1128 : Aborted: Operation received an exception:Status: 3, message: could not create a dilated convolution forward descriptor, in file tensorflow/core/kernels/mkl_conv_ops.cc:1125
Traceback (most recent call last):
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1334, in _do_call
return fn(*args)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1319, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1407, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.AbortedError: Operation received an exception:Status: 3, message: could not create a dilated convolution forward descriptor, in file tensorflow/core/kernels/mkl_conv_ops.cc:1125
[[{{node Conv2D}}]]
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "a.py", line 15, in <module>
{x: np.random.rand(1, 3, 600, 800)})
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 929, in run
run_metadata_ptr)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1152, in _run
feed_dict_tensor, options, run_metadata)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1328, in _do_run
run_metadata)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1348, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.AbortedError: Operation received an exception:Status: 3, message: could not create a dilated convolution forward descriptor, in file tensorflow/core/kernels/mkl_conv_ops.cc:1125
[[node Conv2D (defined at a.py:10) ]]
Errors may have originated from an input operation.
Input Source operations connected to node Conv2D:
random_normal/stddev (defined at a.py:9)
Placeholder (defined at a.py:7)
Pad/paddings (defined at a.py:8)
Original stack trace for 'Conv2D':
File "a.py", line 10, in <module>
out = tf.nn.conv2d(xpad, W, [1, 1, 2, 2], padding="VALID", data_format='NCHW')
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/ops/gen_nn_ops.py", line 1026, in conv2d
data_format=data_format, dilations=dilations, name=name)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
op_def=op_def)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/util/deprecation.py", line 501, in new_func
return func(*args, **kwargs)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3300, in create_op
op_def=op_def)
File "/HOME/.local/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1801, in __init__
self._traceback = tf_stack.extract_stack()
The code will not throw if the padding is [2, 2] instead of [2, 3].
System information
Describe the current behavior
This simple code snippet is legal and runs well on a GPU. However, when run on a CPU with MKL build it throws:
The code will not throw if the padding is [2, 2] instead of [2, 3].