train_all.py: for training models
test.py: for testing
encode.py: defining encoding and decoding functions for testing
util.py: defining network layers
output_interal_value.py: functions for deriving RD loss and coefficients of image samples
clustering_p.py: functions for deriving histogram of coefficients and cluster histograms for multi-probability model
a pretrained demo ensemble model is provided on another branch "model"