Add universal layer spec, custom features - Githubissues

OverLordGoldDragon / see-rnn

RNN and general weights, gradients, & activations visualization in Keras & TensorFlow

MIT License

179 stars 21 forks source link

Add universal layer spec, custom features #16

Closed OverLordGoldDragon closed 4 years ago

OverLordGoldDragon commented 4 years ago

Features:

Universal layer specifier, _id, that is a layer name, layer index, or list of either or of both
Support for figure saving on all visuals
rnn_histogram and rnn_heatmap now plot bias w/ kernel & recurrent kernel in one figure
All plots now return fig and axes objects

Breaking:

layer_name & layer_idx deprecated; _id now takes place of both, and can be a list containing both
label_channels in features_1D deprecated in favor of annotations

Misc:

Added tests
Updated docstrings
K_eval removed from utils.py (unused)
_filter_duplicates_by_keys added to utils.py
_save_rnn_figs added to utils.py