[GitHub] regzhuce commented on issue #7375: Can I set instance weight when training?

2017-09-28 Thread git
regzhuce commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-332829637 Say, I got lots of samples, but not all samples are the same importance. I wanna to give every sample an importance, i.e. instance

[GitHub] regzhuce commented on issue #7375: Can I set instance weight when training?

2017-08-10 Thread git
regzhuce commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-321723588 @thirdwing Thanks Any proposals for my problem? This is an

[GitHub] regzhuce commented on issue #7375: Can I set instance weight when training?

2017-08-07 Thread git
regzhuce commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-320857766 It's so strange when predicting, I have to feed a constant weight to the model, and get a loss value but the prediction value.