search

traja-team / traja

Python tools for spatial trajectory and time-series data analysis
https://traja.readthedocs.io
MIT License
98 stars 25 forks source link

Unused variable #65

Open JustinShenk opened 3 years ago

JustinShenk commented 3 years ago

https://github.com/traja-team/traja/blob/e3dcf25d69a08e3f58844e96d55bc1ca9917c794/traja/models/visualizer.py#L64

Please remove unused variable self_connections, and in the future lint before committing code.

Saran-nns commented 3 years ago

Yes, self connections/time-delayed connections are not used in the current latent visualization. Also, note this visualizer support only recurrent networks. At the moment, latent models in traja are feedforward. However, the encoders and decoders have LSTM layer's whose activations/connections can be visualized by passing their hidden to hidden weights to the wrapper.