This project uses SVCCA to measure the similarity between layers in HAR models
The datasets in this project are Opportunity, pamap2 and daphnet gait. The links are below
Opp: http://archive.ics.uci.edu/ml/datasets/OPPORTUNITY+Activity+Recognition
pamap2: http://archive.ics.uci.edu/ml/datasets/PAMAP2+Physical+Activity+Monitoring
daphnet gait: http://archive.ics.uci.edu/ml/datasets/Daphnet+Freezing+of+Gait
Once downloaded the datasets, please save the dataset in the data source file.
The whole project is run on Colab with GPU acceleration Please notice the default data path is in the colab environment
Once downloaded the data, used the code in demo.ipynb to train and test the model.