Andrej

I recently implemented beam search for an RNN Language Model in context of Image Captioning in NeuralTalk2 repo (https://github.com/karpathy/neuraltalk2/blob/master/misc/LanguageModel.lua#L166). It does makes things work quite a bit better there. I...

Thanks! This looks great and I'd be happy to merge something like this when the kinks are ironed out.

Is it easy to also include, e.g. validation loss? looking at it in context of training loss is very useful usually. We can expand on this in the future I...

Ok I had a closer look at the code and while I am onboard with the general idea of including web-based visualization of the training progress, I am hesitant to...

I was just thinking about this as well. @soumith is the preferred solution to always save CPU models and explicitly convert to GPU int he sampling script if the user...

@soumith I'm not fully comfortable with some of these API and best practices. I'm planning to iterate over all entries in proto, convert them with `:float()`, save to file, and...

@soumith ahhh! Glad I asked, that's precisely the kind of gotcha I was afraid of. I'll keep this in mind.

Yes I think I was going to do this but then decided it would be tricky due to parameter tying issues. The problem is that when you case the model...

I created a quick script to convert char-rnn GPU models to CPU models as a temporary solution to this issue. In the long run we'll want to always save a...

@soumith Hey Soumith RE: this issue with char-rnn, I think there is support now in Torch that doesn't destroy parameter sharing when model is shipped between CPU GPU. Though I'm...