search

AdityaAsopa / eiDynamics

0 stars 1 forks source link

things to add #2

Open AdityaAsopa opened 3 years ago

AdityaAsopa commented 3 years ago

  1. Meta analysis

    • Single pickle file,
    • Peak calculation,
    • Low pass filtering
    • Save or not

  2. Neural network

    • Slice out sweeps,
    • Attach parameters to every sweep
    • Design an NN (CNN?)

  3. Ra, IR, baseline flag filtering

  4. 1sq histogram for pulses

AdityaAsopa commented 2 years ago

Done: Peak Calculation sanity check, Filtering, Saving, Single pickle file and h5 file with raw traces for every cell. Tau, IR flag for every sweep in h5 and excel file, Attached parameters like E/I, numSquares, IR, Tau, stimFreq, pulseWidth etc for every sweep in excel file (pulse peak responses), and h5 file (raw sweeps)

Still remaining:

  1. Histograms,
  2. function to normalize responses to first pulse or test pulse response