small io improvements - Githubissues

tensorflow / lucid

A collection of infrastructure and tools for research in neural network interpretability.

Apache License 2.0

4.65k stars 655 forks source link

Closed michaelpetrov closed 4 years ago

michaelpetrov commented 4 years ago

A few small improvements to make io scopes a bit more flexible and improve inference of data formats by allowing deeper recursion.

ludwigschubert commented 4 years ago

LGTM!