Michael Oliver

Making solvers more general

7

I'm working on an implementation of RMSProp for Mocha, but I quickly ran into the issue of it needing more parameters than SGD (to control things like the exponential window,...

Use gpu for random number generation

1

Thanks for posting this. However I found you can make things much faster by using the gpu for generating your random numbers as below: ```python class GaussianDropout(nn.Module): def __init__(self, alpha=1.0):...

Update .travis.yml

1

Update for consistency with keras and keras-contrib

Small generalization of your code to allow the functions to which constraints are applied to be able to take arguments. This allows for using multiple loss functions, constraining a model's...