Skip to content
This repository has been archived by the owner on Nov 25, 2019. It is now read-only.

Latest commit

 

History

History
33 lines (25 loc) · 1.86 KB

README.md

File metadata and controls

33 lines (25 loc) · 1.86 KB

Clustering neuron from allen db

  • 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 These codes are no longer maintained and very dirty.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.

Data pipeline

  1. preparing raw data from allen brain institute using AllenSDK notebook
    1. preparing new raw data for revising manuscript. notebook
  2. Visualizing raw data using density plot notebook, just verification
  3. Dividing data into train, test set in rmd
    1. for revising manuscript rmd

LASSO, RF done by incheol

  1. For binary classification in rmd, excitatory line classification in rmd, and inhibitory line classification in rmd.
  2. models saved in ./lasso_rf/R_models/
  3. reload models test in rmd

ANN learning pipeline

  1. data processing for tensorflow learning from R data(incheol) notebook
    1. one-hot coding
    2. minmax scaling
  2. For ANN, view in the folder named ANN
  3. coarse searching hyperparameter(learning rate and L2 beta) = ./ANN/NO_1_output_input_coarse_searching.py
  4. fine searching = ./ANN/NO_2_output_input_fine_searching.py
  5. top 10 model tensorboard logging and model saving = ./ANN/NO_3_output_input_logging.py
  6. selection top model by inspecting tensorboard log (./ANN/logs/output_input/)
  7. top model restore and choosing best epoch step, saving results = ./ANN/NO_4_output_input_restore.py
  8. all of the final results in ./ANN/results/