I am trying to used SMOTE and Feature Selection by following this paper http://jad.shahroodut.ac.ir/article_825_679b8f128dec2874a8fbc314fc922127.pdf
In this paper, the authors have mentioned about 4 scenarios. I took 2 scenarios as example.
- Scenario 3 (S3): using CFS, we select features from the sampled data, and create the training dataset based on the original data.
- Scenario 4 (S4): using CFS, we select features from the sampled data, and create the training dataset based on the sampled data.
The codes for the scenarios (data sampling and feature selection) are:
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0) sm = SMOTE(random_state=2) X_train_res, y_train_res = sm.fit_sample(X_train, y_train.ravel()) select = SelectFromModel(LogisticRegression(class_weight='balanced',penalty="l1",C=0.01,solver='liblinear')) X_ = select.fit_transform(X_train_res, y_train_res) model = LogisticRegression(class_weight='balanced',penalty='l2',C=0.01).fit(X_,y_train_res) y_preds = model.predict(X_)
My question is how could I create training data based on the samples or the original data as mentioned in the scenarios?