Home
Description: WeightWatcher is an open source data free diagnostic tool for training, monitoring and improving Deep Learning models.
WeightWatcher (w|w) is an open-source, diagnostic tool for analyzing Deep Neural Networks (DNN), without needing access to training or even test data. It is based on theoretical research into Why Deep Learning Works, using the new Theory of Heavy-Tailed Self-Regularization (HT-SR), published in JMLR and Nature Communications .
Easy to use WeightWatcher requires just a few lines of code to generate layer-by-layer diagnostics for your Deep Learning models. It supports most Tensorflow/Keras, pyTorch, and HuggingFace CV and NLP/Transformer models (Dense and Conv2D layers).
The weightwatcher HTSR theory tells us if and when a specific DNN layer has converged properly; it is unique in this regard and the only theory capable of this. When running watcher.analyze() , you will obtain a pandas dataframe containing several layer quality metrics. In particular, the weightwatcher alpha metric can tell us if a layer has is well trained or not. Specifically, the layer quality metric alpha should be between 2 and 6. Here we have run weightwatcher on 2 of the currently newly popular Bloom
Home