But the logic here does not work for float64. If you want to support float64, it is probably better to have a specific logic block. For now, this logic literally means "if DType is not float32, do this..." which is incorrect.
[ Full content available at: https://github.com/apache/incubator-mxnet/pull/12261 ] This message was relayed via gitbox.apache.org for [email protected]
