Kyle Mathewson

http://kylemathewson.com [email protected]

University of Alberta Edmonton, AB Associate Professor of Psychology and Neuroscience, Director of the Attention Perception and Performance Lab

Results 28 issues of


                                            Kyle Mathewson

MUSE Load - Allow different number of sessions per subject (array of number of sessions), and allow for more than 2 sessions loaded

Scaling model

you will want to fit a scaler (mean / standard deviation) https://scikit-learn.org/stable/modules/preprocessing.html --- on the training data, and then use that scaling model on the test/validation Because you will not...

Might want to use stratify on the train_test_split

May want to try this: glorot_normal https://keras.io/initializers/ for the intiialization

get tensorboard working again

1

comment

And you can implement early_stopping as well, to make sure that your model doesn't start getting worse on the validation:

https://keras.io/callbacks/#earlystopping

How does number of features relate to number of data points relate to number of parameters?

Should start using logging instead of print, it will help in the long run.

worth trying xgboost on this problem, it is super fast/ tree based (descriptive) and powerful

Unbalanced classes. Stratified sampling. k-fold stratified cross validation helps

1
2
3
›