[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281234142 ## File path: docs/tutorials/mkldnn/MKLDNN_README.md ## @@ -19,7 +19,9 @@ A better training and infe

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281232336 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,83 @@ + + + + + + + + + + + + + + +

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281230120 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,83 @@ + + + + + + + + + + + + + + +

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281230120 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,83 @@ + + + + + + + + + + + + + + +

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281222337 ## File path: docs/install/index.md ## @@ -187,7 +187,12 @@ $ pip install mxnet --pre -MXNet of

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281222337 ## File path: docs/install/index.md ## @@ -187,7 +187,12 @@ $ pip install mxnet --pre -MXNet of

[GitHub] [incubator-mxnet] xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

xinyu-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281222337 ## File path: docs/install/index.md ## @@ -187,7 +187,12 @@ $ pip install mxnet --pre -MXNet of

[GitHub] [incubator-mxnet] juliusshufan opened a new pull request #14893: Integrating the MKL VML functions to MXNET to speed-up the mathematic computation

juliusshufan opened a new pull request #14893: Integrating the MKL VML functions to MXNET to speed-up the mathematic computation URL: https://github.com/apache/incubator-mxnet/pull/14893 ## Description ## Intel MKL provides a wide range of the generic vectored mathematic (VML) functions

[incubator-mxnet] branch master updated (f6ef206 -> 42ede50)

This is an automated email from the ASF dual-hosted git repository. arcadiaphy pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from f6ef206 [Bugfix] Fix layer norm for large input shape (#14870) add 42ede50 rewrite test_custom_op

[GitHub] [incubator-mxnet] arcadiaphy merged pull request #14878: fix flaky test test_custom_op_exc

arcadiaphy merged pull request #14878: fix flaky test test_custom_op_exc URL: https://github.com/apache/incubator-mxnet/pull/14878 This is an automated message from the Apache Git Service. To respond to the message, please lo

[GitHub] [incubator-mxnet] TaoLv commented on issue #14891: [Doc] Add MKL-DNN operator list

TaoLv commented on issue #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#issuecomment-489618602 @xinyu-intel @ciyongch please help to review. Thank you. This is an automated me

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 9f1bcf7 Bump the publish ti

[GitHub] [incubator-mxnet] chinakook commented on issue #13925: [Feature Request] Add Simple Recurrent Unit (SRU)

chinakook commented on issue #13925: [Feature Request] Add Simple Recurrent Unit (SRU) URL: https://github.com/apache/incubator-mxnet/issues/13925#issuecomment-489610913 These bar charts looks nice, but SRU need more layers to catch the accuracy of LSTM. In total, the speed would be near

[GitHub] [incubator-mxnet] AnaRhisT94 edited a comment on issue #14881: [Solved]MXNet to ONNX export bug with "out_shape" when using Crop

AnaRhisT94 edited a comment on issue #14881: [Solved]MXNet to ONNX export bug with "out_shape" when using Crop URL: https://github.com/apache/incubator-mxnet/issues/14881#issuecomment-489582793 Just used Crop with num_args = 1 and sliced the h_w accordingly in the .json file (Directly).

[GitHub] [incubator-mxnet] AnaRhisT94 edited a comment on issue #14881: [Solved]MXNet to ONNX export bug with "out_shape" when using Crop

AnaRhisT94 edited a comment on issue #14881: [Solved]MXNet to ONNX export bug with "out_shape" when using Crop URL: https://github.com/apache/incubator-mxnet/issues/14881#issuecomment-489582793 Just used Crop with num_args = 1 and sliced the h_w accordingly in the .json file (Directly) -

[GitHub] [incubator-mxnet] AnaRhisT94 commented on issue #14875: MXNet to ONNX export bug

AnaRhisT94 commented on issue #14875: MXNet to ONNX export bug URL: https://github.com/apache/incubator-mxnet/issues/14875#issuecomment-489583284 @ehsanmok mxnet isn't responsible for this error, it's only ONNX. update your ONNX version, should work when ONNX==1.2.2

[GitHub] [incubator-mxnet] AnaRhisT94 closed issue #14880: Changing a Crop to Slice_axis for ONNX conversion

AnaRhisT94 closed issue #14880: Changing a Crop to Slice_axis for ONNX conversion URL: https://github.com/apache/incubator-mxnet/issues/14880 This is an automated message from the Apache Git Service. To respond to the messag

[GitHub] [incubator-mxnet] AnaRhisT94 closed issue #14881: MXNet to ONNX export bug with "out_shape" when using Crop

AnaRhisT94 closed issue #14881: MXNet to ONNX export bug with "out_shape" when using Crop URL: https://github.com/apache/incubator-mxnet/issues/14881 This is an automated message from the Apache Git Service. To respond to th

[GitHub] [incubator-mxnet] AnaRhisT94 commented on issue #14880: Changing a Crop to Slice_axis for ONNX conversion

AnaRhisT94 commented on issue #14880: Changing a Crop to Slice_axis for ONNX conversion URL: https://github.com/apache/incubator-mxnet/issues/14880#issuecomment-489582895 **Solved in ONNX==1.2.2** This is an automated messag

[GitHub] [incubator-mxnet] AnaRhisT94 commented on issue #14881: MXNet to ONNX export bug with "out_shape" when using Crop

AnaRhisT94 commented on issue #14881: MXNet to ONNX export bug with "out_shape" when using Crop URL: https://github.com/apache/incubator-mxnet/issues/14881#issuecomment-489582793 **Solved in ONNX==1.2.2** This is an automate

[GitHub] [incubator-mxnet] ymjiang commented on issue #14485: Any suggestion to accelerate parameter update on PS for distributed training?

ymjiang commented on issue #14485: Any suggestion to accelerate parameter update on PS for distributed training? URL: https://github.com/apache/incubator-mxnet/issues/14485#issuecomment-489579807 @eric-haibin-lin For a very simplistic case with 1 worker (one 1080Ti GPU) & 1 PS and train V

[GitHub] [incubator-mxnet] songziqin opened a new issue #14892: embedding_lookup_sparse(wide_w, ids, weights, combiner="sum") C++ api

songziqin opened a new issue #14892: embedding_lookup_sparse(wide_w, ids, weights, combiner="sum") C++ api URL: https://github.com/apache/incubator-mxnet/issues/14892 I have read the tensorflow phython code embedding_lookup_sparse(wide_w, ids, weights, combiner="sum") . and i wonder if the

[GitHub] [incubator-mxnet] rongzha1 commented on issue #14889: fix add_n bug: when input mem overlap with output mem, results is wrong

rongzha1 commented on issue #14889: fix add_n bug: when input mem overlap with output mem, results is wrong URL: https://github.com/apache/incubator-mxnet/pull/14889#issuecomment-489547986 > Nice catch! Would you mind adding an unit test to test_ndarray.py? e.g. > > ``` > x = mx.n

[GitHub] [incubator-mxnet] pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281086615 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,85 @@ + + + + + + + + + + + + + +

[GitHub] [incubator-mxnet] pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281086985 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,85 @@ + + + + + + + + + + + + + +

[GitHub] [incubator-mxnet] pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list

pengzhao-intel commented on a change in pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#discussion_r281088098 ## File path: docs/tutorials/mkldnn/operator_list.md ## @@ -0,0 +1,85 @@ + + + + + + + + + + + + + +

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 75b04e2 Bump the publish ti

[GitHub] [incubator-mxnet] TaoLv commented on issue #14888: Memory usage estimator

TaoLv commented on issue #14888: Memory usage estimator URL: https://github.com/apache/incubator-mxnet/issues/14888#issuecomment-489521685 Curious to know why we need an estimator. Can a estimated number stop the users from trying a larger input size? I think they would not stop until they

[GitHub] [incubator-mxnet] haojin2 edited a comment on issue #14842: Implementation for equivalence of tf.moments

haojin2 edited a comment on issue #14842: Implementation for equivalence of tf.moments URL: https://github.com/apache/incubator-mxnet/pull/14842#issuecomment-489521394 @eric-haibin-lin The op gives variance of the input on arbitrary axis/axes, so it's technically the `np.var` operator's eq

[GitHub] [incubator-mxnet] haojin2 commented on issue #14842: Implementation for equivalence of tf.moments

haojin2 commented on issue #14842: Implementation for equivalence of tf.moments URL: https://github.com/apache/incubator-mxnet/pull/14842#issuecomment-489521394 @eric-haibin-lin The op gives variance of the input on arbitrary axis/axes, so it's technically the `np.var` operator's equivalenc

[GitHub] [incubator-mxnet] TaoLv commented on issue #14891: [Doc] Add MKL-DNN operator list

TaoLv commented on issue #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891#issuecomment-489518409 @pengzhao-intel please review~ This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] TaoLv opened a new pull request #14891: [Doc] Add MKL-DNN operator list

TaoLv opened a new pull request #14891: [Doc] Add MKL-DNN operator list URL: https://github.com/apache/incubator-mxnet/pull/14891 ## Description ## Add MKL-DNN optimized operator list and graph fusion pattern list. This PR is part of the plan in https://github.com/apache/incubator

<    1   2