search

Ma-Xinzhi / LightConvNet

MIT License
18 stars 4 forks source link

Input data #2

Open YamengGu opened 5 months ago

YamengGu commented 5 months ago

I am really interested in this model. Wonder if you could upload the input data the model used. Or describe how the input data was organized in the .npy file.

Thanks.

Ma-Xinzhi commented 4 months ago

I am really interested in this model. Wonder if you could upload the input data the model used. Or describe how the input data was organized in the .npy file.

Thanks.

The data processing file can be found in the folder 'data'. 'preprocess_*.py' is used to preprocess and save the raw data in the .npy file. The shape of saved raw data in the .npy file is [N, C, T], where N represents the number of trials, C represents the number of channels, T represents the number of sample points. Then, 'save_multifreq_data.py' is used to generate multi-view data filtered by a filter bank. The shape of multi-view data is [N, F, C, T], where F represents the number of filters. Finally, the multi-view data is forwarded as the input data to the proposed network. Related details can also be found in the paper.