Sasha Goodman

feature request: dropout during prediction phase

4

Yarin Gal's thesis demonstrated that by allowing dropout during prediction, not just training, a set predictions can be produced and then analyzed to reveal model uncertainty (the quantiles of the...

Why does this model not run in mobile Chrome and mobile Safari?

3

Just made a toy example of an insult dector using a character level CNN: https://keras-predict-r.c9users.io/insults.html The F1 score on the validation set is around 0.65, so it's not great but...

Sparse matrices

1

Given the limited memory and bandwidth in a browser, sparse matrices could speed up transmission times and lower the memory usage. Some example use cases: (1) If a model inefficiently...