Built a Software Defect Prediction Model using an improved Deep Forest algorithm and Autoencoder by Forest. The original input features are transformed by data augmentation method to enhance the ability of feature extraction, and the autoencoder by forest performs data dimensionality reduction on the features
