Federico Vaggi

Return type is tuple of solution and nfe not just solution

3

- https://github.com/jacobjinkelly/easy-neural-ode/blob/1ae0836959fe61d11b09540369f3dcd230e58565/lib/ode.py#L554-558 The return type is actually a tuple of the solution and the number of function evaluations: https://github.com/jacobjinkelly/easy-neural-ode/blob/1ae0836959fe61d11b09540369f3dcd230e58565/lib/ode.py#L747 Super easy to clean up, I just thought it was some...

WIP Analytical Derivation of Sensitivity Equations

13

There are three major ways of calculating the time-dependent sensitivity of a system of differential equations: - Automatic Differentiation - Finite Differences - Sensitivity Equations Unlike FD or AD, sensitivity...

Predicted ESM2 logits depend on other elements within a batch

11

Thank you so much for all the work on ESM and ESM2. I ran into some surprising behaviour: **Bug description** ESM2 predicts slightly different logits even when in eval mode...

Autocompetition not working on Windows 7/8

13

Hi, I was pretty excited to try out Sublime Rope, but somehow it does not work on Windows at all. I am guessing this is because Sublime Rope thinks that...

In very large datasets, encoding the categorical variables often dominates the preprocessing time compared to scaling numerical columns. In that case, it's often faster to parallelize across rows rather than...

If you have a column with missing values, then, the nan values will cause an error in the minhash encoder in this step: string = spaces + " ".join(string.lower().split()) +...

https://github.com/pcerda/string_categorical_encoders/blob/406986d0a8b6e6286245fdd661605ccbc92dc806/column_encoder.py#L19 this causes an import error if fasttext is installed with pip. Is this not the standard facebook fasttext? I think this is what happens if you install the git...

I did a few very minor code clean ups to insure that the example can be run under Python3.