Open roywei opened 5 years ago
after https://github.com/apache/incubator-mxnet/pull/14661 and https://github.com/apache/incubator-mxnet/pull/14869. Softmax operator has more strict partial shape inference. This will result in the ndim() of softmax input maybe unknown during dimension check. It will be inferred at a later time.
I have created PR for the case of ndim == 0 in for mxnet in https://github.com/awslabs/keras-apache-mxnet/pull/232. I'm keeping this issue open to see if we can find alternatives of modifying Keras front end code.
ndim == 0
after https://github.com/apache/incubator-mxnet/pull/14661 and https://github.com/apache/incubator-mxnet/pull/14869. Softmax operator has more strict partial shape inference. This will result in the ndim() of softmax input maybe unknown during dimension check. It will be inferred at a later time.
I have created PR for the case of
ndim == 0in for mxnet in https://github.com/awslabs/keras-apache-mxnet/pull/232. I'm keeping this issue open to see if we can find alternatives of modifying Keras front end code.