- for 2nd paper using allen db, clustering neurons with MLs, collaborating incheol.
- our paper was published at Brain Research Bulletin
- Caution!! These codes are no longer maintained and very dirty. Be careful.
- preparing raw data from allen brain institute using AllenSDK notebook
- preparing new raw data for revising manuscript. notebook
- Visualizing raw data using density plot notebook, just verification
- Dividing data into train, test set in rmd
- for revising manuscript rmd
- For binary classification in rmd, excitatory line classification in rmd, and inhibitory line classification in rmd.
- models saved in ./lasso_rf/R_models/
- reload models test in rmd
- data processing for tensorflow learning from R data(incheol) notebook
- one-hot coding
- minmax scaling
- For ANN, view in the folder named ANN
- coarse searching hyperparameter(learning rate and L2 beta) = ./ANN/NO_1_output_input_coarse_searching.py
- fine searching = ./ANN/NO_2_output_input_fine_searching.py
- top 10 model tensorboard logging and model saving = ./ANN/NO_3_output_input_logging.py
- selection top model by inspecting tensorboard log (./ANN/logs/output_input/)
- top model restore and choosing best epoch step, saving results = ./ANN/NO_4_output_input_restore.py
- all of the final results in ./ANN/results/