search

neil-gallagher / CSFA

Codebase for Cross-Spectral Factor Analysis (Gallagher et al., 2017)
12 stars 6 forks source link

Prediction score of each electome factor #3

Open DanielStrahnen opened 3 years ago

DanielStrahnen commented 3 years ago

Thank you for the nice repository! I would like to ask about the prediction scores of each electome factor: As I understand it, the scores that are saved at the end of trainCSFA represent the activity of each electome factor in each trial. But is there a way to rank them regarding their predictability (i.e., the higher the score the more important for classification?). Basically my goal would be to get to the same figure as in Hultman et al., Cell, 2018 in Figure 1F to get a ranking of factors regarding their contribution to classifying between two conditions.

neil-gallagher commented 3 years ago

Those results were achieved using a supervised dCSFA model with a linear SVM classifier. The 'Prediction Score' in that figure was simply the linear weight assigned to that factor by the SVM, after normalizing the weight vector so that all weights summed to 1.