connorgoggins commented on issue #17456: Implement remaining nn_basic ops in
opperf
URL: https://github.com/apache/incubator-mxnet/pull/17456#issuecomment-580513915
[Group of operator test - all NN Basic ops
(GPU)](https://gist.github.com/connorgoggins/671d248f09414400836665046d365627)
[Group of operator test - all NN Basic ops
(CPU)](https://gist.github.com/connorgoggins/d683e7ca75135bf2bf96df992d98b23c)
[Full OpPerf test (GPU)*
](https://gist.github.com/connorgoggins/2d4d2ff6dca61494eb8151a5106fec6c)
[Full OpPerf test
(CPU)](https://gist.github.com/connorgoggins/e7028106ff8fc9128301826855a70973)
*Note: couldn't run SoftmaxOutput op backwards on GPU - see [previously
documented issue](https://github.com/apache/incubator-mxnet/issues/880) and
couldn't run im2col either forwards or backwards on GPU - see [new documented
issue](https://github.com/apache/incubator-mxnet/issues/17493). In both cases,
simply switching the context from cpu to gpu leads the op to fail.
Also couldn't run the topk op in the full test on GPU because of a
[documented issue](https://github.com/apache/incubator-mxnet/issues/15703)
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:
us...@infra.apache.org
With regards,
Apache Git Services