NRauschmayr commented on issue #13660: A question about the mechanism of autograd URL: https://github.com/apache/incubator-mxnet/issues/13660#issuecomment-449026065 The model parameters get only updated once you call optimizer.step(). The easiest would be to just feed all the data once e.g. ``` with autograd.record(True): network_out = model(mx.nd.array[[input1],[input2]]) loss = criterion(network_out, mx.nd.array[[gt],[gt2]]) autograd.backward(loss) optimizer.step() ```
---------------------------------------------------------------- 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: [email protected] With regards, Apache Git Services
