apeforest commented on issue #16218: Improving performance of argmax operator URL: https://github.com/apache/incubator-mxnet/pull/16218#issuecomment-534312819 @drivanov Thanks a lot for your contribution and profiling the performance. This looks really great. We'd love to merge your PR ASAP. I just had a few comments regarding large tensor support. Since we plan to support large tensor (index over MAX_INT32), could you please verify if you change can run through a large array by compiling MXNet with compiler flag USE_INT64_TENSOR_SIZE on? We've already had a nightly test for this operator: https://github.com/apache/incubator-mxnet/blob/master/tests/nightly/test_large_array.py#L1031 Thanks!
---------------------------------------------------------------- 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
