You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
- Added preprocessing script (`preprocess.py`) to convert CSV or EDF files to NumPy .npy files with various preprocessing steps including notch filtering and bandpass filtering.
7
+
- Implemented parallel processing for preprocessing using the `--parallel` flag.
8
+
- Added support for TUH EEG files in preprocessing.
9
+
- Included detailed README with instructions for using `preprocess.py` and example usage for Crown CSV files and TUH EEG files.
10
+
- Integrated `wandb` for experiment tracking.
11
+
- Added `CSVLogCallback` class in `src/trainer/make.py` for logging training and evaluation metrics to CSV files.
12
+
- Provided `train_parallel.sh` script for distributed training using PyTorch with multiple GPUs.
13
+
- Included dependencies in `requirements.txt` and `requirements-dev.txt` for easy setup.
14
+
15
+
### Documentation
16
+
- Comprehensive README with setup instructions, preprocessing details, and example usage.
17
+
- Links to external resources for additional help with tools like `tmux`.
18
+
19
+
### References
20
+
- Based on NeuroGPT by Wenhui Cui et al.
21
+
- Inspired by Neurosity Foundational Model by Jeremy Nixon and AJ Keller.
0 commit comments