/blechRNN

Estimating neural firing rates using LSTM

Primary LanguagePython

BlechRNN

Pipeline to estimate neural firing rates using an autoregressive LSTM.

Modules:

  • 'get_data': Load data from a specified file and preprocess it.
  • 'model': Define the LSTM model.
  • 'train': Train the LSTM model and predict neural firing rates.
  • 'output': Save the predicted neural firing rates to a specified file, save the model to a specified file, and generate plots

Functionality roadmap:

  • Ability to run prespecified model on a single dataset
  • Batch processing of multiple datasets
  • Hyperparameter optimization
  • Remote execution