yemenr / machine-learning-toolbox

Machine-Learning-Toolbox

1.DataToolBox

(1) Metric:

acc=get_acc(real_label,predict_label)

auc=get_auc(real_label, scores)

[tpr,tnr,macc]=get_macc(real_label, predict_label)

(2) Re-sampling:

[data_much,data_less,much_label,less_label]=divide_data(data, label)

Random Under-Sampling: [train_data_temp,train_label_temp]=RUS(data_much, data_less, much_label, less_label)

Random Over-Sampling: [train_data_temp,train_label_temp]=ROS(data_much, data_less, much_label, less_label)

2.Ensemble

(1) Stacking

stacking(clfs,X_train,y,X_test,nfolds=5,stage=1,random_seed=2017,shuffle=True,clfs_name=None,final_clf=None)

3. MatMHKS

A matrix based linear classifier

clf=MatMHKS(penalty='l2', C=1.0, matrix_type=None,class_weight=None, max_iter=100,u0=0.5,b0=10**(-6),eta=0.99,min_step=0.0001,multi_class='ovr', verbose=0)

clf.fit(X,y)

clf.predict(X)

clf.predict_proba(X)

4.generator

generator for keras/tensorflow (enhanced,imbalanced data)

5.metrics

tpr tnr for keras/tensorflow

6.loss

(1) focal loss

(2) center loss

(3) triplet loss

(4) island loss