You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/05/08 01:01:12 UTC

[GitHub] ashokei commented on a change in pull request #10591: [MXNET-365] handle inplace in mkldnn FallBackCompute

ashokei commented on a change in pull request #10591: [MXNET-365] handle inplace in mkldnn FallBackCompute
URL: https://github.com/apache/incubator-mxnet/pull/10591#discussion_r186592725
 
 

 ##########
 File path: tests/python/mkl/test_mkldnn.py
 ##########
 @@ -121,47 +100,33 @@ def __getitem__(self, key):
 
     X = (32, 3, 32, 32)
     # trigger mkldnn execution thread
-    y = net(nd.array(np.ones(X))).asnumpy()
+    y = net(mx.nd.array(np.ones(X))).asnumpy()
 
     # Use Gluon dataloader to trigger different thread.
     # below line triggers different execution thread
     for _ in loader:
-        y = net(nd.array(np.ones(X))).asnumpy()
+        y = net(mx.nd.array(np.ones(X))).asnumpy()
         # output should have 0.3376348
-        assert_almost_equal(y[0, 0, 0, 0], 0.3376348)
+        assert_almost_equal(y[0, 0, 0, 0], 0.016711406)
 
 Review comment:
   it just computes with fixed 1s as input, and this is the value i see as a result with GPU or cpu (non-mkldnn mode).

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services