reminisce commented on issue #9007: float16 argmax breaks on negative inputs URL: https://github.com/apache/incubator-mxnet/issues/9007#issuecomment-350529028 Unary reduce ops have the problem of handling float16 correctly. For example, ```python a = mx.nd.array([-2, 0], dtype=np.float16) print(mx.nd.max(a)) [ 6.10351562e-05] <NDArray 1 @cpu(0)> ``` I'm guessing it might be related to setting initial minimum value for float16 type. https://github.com/dmlc/mshadow/blob/2d7780c3f2eefe4453fa419862d1b2089bedb8d5/mshadow/extension/reduce_with_axis.h#L121
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on 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