optima2005 commented on a change in pull request #4604: [FRONTEND][TF] Add
conv3d
URL: https://github.com/apache/incubator-tvm/pull/4604#discussion_r362297504
##########
File path: tests/python/frontend/tensorflow/test_forward.py
##########
@@ -469,6 +472,57 @@ def test_forward_convolution():
'NHWC', [1, 8, 8, 1])
+#######################################################################
+# Convolution3D
+# -----------
+
+
+def _test_convolution3d(opname, tensor_in_sizes, filter_in_sizes,
+ dilations, strides, padding, data_format,
+ deconv_output_shape=[]):
+ """ One iteration of 3D convolution with given shapes and attributes """
+
+ total_size_1 = np.prod(tensor_in_sizes)
+ total_size_2 = np.prod(filter_in_sizes)
+ # Initializes the input tensor with array containing incrementing
+ # numbers from 1.
+ data_array = [f * 1.0 for f in range(1, total_size_1 + 1)]
+ filter_array = [f * 1.0 for f in range(1, total_size_2 + 1)]
+
+ with tf.Graph().as_default():
+ in_data = array_ops.placeholder(shape=tensor_in_sizes, dtype='float32')
+ in_filter = constant_op.constant(
+ filter_array, shape=filter_in_sizes, dtype='float32')
+ if data_format == 'NDHWC':
+ strides = [1] + strides + [1]
+ dilations = [1] + dilations + [1]
+ else:
+ strides = [1, 1] + strides
+ dilations = [1, 1] + dilations
+
+ if opname == 'conv':
+ nn_ops.conv3d(in_data,
+ in_filter,
+ strides=strides,
+ dilations=dilations,
+ padding=padding,
+ data_format=data_format)
+
+ compare_tf_with_tvm(np.reshape(data_array,
tensor_in_sizes).astype('float32'),
+ 'Placeholder:0', 'Conv3D:0', assert_rtol=1e-2,
cuda_layout="NCDHW")
Review comment:
I switched it high during debug but forgot to recover it. Sorry about that.
I have checked that the default value works for additional tests.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
