search

neuropsychology / NeuroKit.py

A Python Toolbox for Statistics and Neurophysiological Signal Processing (EEG, EDA, ECG, EMG...).
http://neurokit.rtfd.io
MIT License
365 stars 102 forks source link

Mse fix #58

Closed DominiqueMakowski closed 6 years ago

DominiqueMakowski commented 6 years ago

Follow up of #57 Fixing multiscale entropy

DominiqueMakowski commented 6 years ago

Beside fixing testing, I've also added some minor changes:

1) Renaming the MSE parameters to "m" and "r" so it is closer to what is in the papers (I feel this way the transition from litterature to code is smoother) 2) I've found in Norris 2008 the use of another point estimate (the sum of MSE values). Since I like to return many indices by default (for serendipity...), I thought the most neat way to output stuff was to return a dict, which keys are described in the docs. 3) I've added a default value based on Costa 2005 for max_scale_factor (20) with automated cuts in case of problem in SampEn calculation.

What do you think of these changes?

gattia commented 6 years ago
  1. Perfect
  2. Sounds good
  3. Sounds good.

I've checked out the commit and everything looks good to me.

DominiqueMakowski commented 6 years ago

It is merged 🎉 thanks @gattia for everything 😸

gattia commented 6 years ago

Awesome! Thanks for including me in the process. I do most of my coding and work independently, so it's great to learn and see how others do things, thanks for maintaining this.

DominiqueMakowski commented 6 years ago

Do not hesitate if you have other suggestions or ideas ;) cheers