added livelossplot python package

[stared]

Hi @fredhohman!

I add https://github.com/stared/livelossplot, an Python package for visualizing metrics during training. It is extensible, so one can add other qualities to track during training (e.g. https://github.com/stared/livelossplot/blob/master/examples/2d_prediction_maps.ipynb for didactic purposes) . Not sure how you approach packages (rather than publications). Though, since there are some things with "publication: web" I give it a try (especially many machine learning visualization tools are primarily/only libraries).

As a side note, I like this PR-based lists, vide http://p.migdal.pl/interactive-machine-learning-list/.

[fredhohman]

Hi @stared!

Thanks for the PR and the other resources! http://p.migdal.pl/interactive-machine-learning-list/ is a great list too!

This repo accompanies our IEEE TVCG paper and focused on (mostly) published interfaces and systems for visual analytics in deep learning; we didn't discuss toolkits too much.

Let me think about this a bit more—maybe we could add a toolkit/library section to include your PR and other resources that aren't explicitly research papers.

Thanks!