Cédric Bovar

It seems that **RegressionLayer** disappeared at some point (from tag 0.3.2). I will try to reintroduce it this week end.

I think I have never implemented _RegressionLayer_ since ConvNetSharp handles batch.

_RegressionLayer_ committed

Have you tried with a lower learning rate? E.g. 0.001

I haven't had time to look at the code yet. But you could maybe make the problem even simpler (like [this](https://medium.com/@awjuliani/super-simple-reinforcement-learning-tutorial-part-1-fd544fab149)) to make it easier to debug.

Do you want to output the two output kinds at the same time ? It's already possible to have proba as output using SoftmaxLayer as last layer (e.g. Mnist demo)...

Do you want to get the output of the softmax layer And the output of the layer just before softmax ? Or do you want to get the output of...

Here is a quick example: (to make it simple, I only don't use any dataset: I use one constant input and two constant outputs) ```c# using System; using System.Collections.Generic; using...

It looks like a duplicate of #13 cuDNN implements batch normalization so GPU part should be straight forward. PR welcome :)

Hi! Currently there is no unpooling or deconvolution. It's something I have planned to do but it's low on my priority list. PR welcome :)