[GitHub] azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False

azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False URL: https://github.com/apache/incubator-mxnet/issues/10994#issuecomment-417379102 docs mention that this is only used for a few operators such as batch_norm

[GitHub] azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False

azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False URL: https://github.com/apache/incubator-mxnet/issues/10994#issuecomment-404974972 I think this might be an issue not specific to mkldnn. built without mkldnn ```

[GitHub] azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False

azai91 commented on issue #10994: MKLDNN fails in the backward computation when forward runs with is_train=False URL: https://github.com/apache/incubator-mxnet/issues/10994#issuecomment-404970514 @zheng-da do we know which operators cause this?