hddm-devs / hddm

HDDM is a python module that implements Hierarchical Bayesian parameter estimation of Drift Diffusion Models (via PyMC).
http://ski.clps.brown.edu/hddm_docs/
Other
261 stars 117 forks source link

Alex dev #75

Closed AlexanderFengler closed 2 years ago

AlexanderFengler commented 3 years ago

First attempt at publishable NN extension

twiecki commented 3 years ago

Well, it's not going to win any coding awards ;) but holy cow, a lot of work went into this. Really cool new features. No way I'll be able to review but I left some stylistic points of feedback that I think should be addressed. Adding tf as a dependency is quite heavy, have you considered making that option for people who want to use this new functionality?

twiecki commented 3 years ago

Did you run black on the full code-base?

AlexanderFengler commented 3 years ago

Ran it on the hddm folder yeah. I thought it doesn't makes sense to just run it once over everything for consistency.

twiecki commented 3 years ago

Odd, I have run it too on the master branch and then tried to point this PR to it but there were conflicts.

twiecki commented 3 years ago

OK, I resolved the conflicts, so you have to pull the changes from this branch.

twiecki commented 3 years ago

Not sure why but I had to resolve conflicts again, so make sure to pull.

AlexanderFengler commented 3 years ago

Hm maybe I am doing something wrong there, but I pulled before pushing again.