Evaluation metrics broken?!
fix mlxtend dependency? add xgboost dependency add imbalance learn dependency <---- fix openml dependency in imbalanced data -> csv interface pca faces plot broken fancyimpute? improve install instructions
text data too slow?
compare my outlier detection with Alex from scipy
hashing vecorizer in out-of-core different text dataset for exercise? consolidate outlier detection here with aml outlier detection notebook slides?
aclImdb remove unsupervised folder
5 out of core? Makes no sense? extra imbalanced data extra missing values
2 pipeline 3 evaluation 4 text data extra dimensionality reduction extra gradient boosting extra deep learning extra outlier detection extra topic modelling
add deep learning with keras to deep learning add gp back in gp exercise impossibly hard box-cox?