I think it is a very common functionality that assigns a element-wise loss weight to
each sample in the batch.
After reading the doc (http://beta.mxnet.io/r/api/mx.symbol.SoftmaxOutput.html) of
mx.sym.SoftmaxOutput I cannot find a solution.
SoftmaxOutput seems to act as a loss function, but indeed it doesn’t output the loss values.
Then I cannot assign weights to the output of
So how can I assign sample-wise weights to the softmax loss using the symbolic API of