zhy1109 / cnn-adaboost-cnn

#MICNN-AdaBoost

Code for the paper 'AdaBoost MICNN: A New Network Framework for Pulsar Candidate Selection'

This model is suitable for multiple images / one-dimensional signals to be input into adaboost at the same time. It is practical and suitable for imbalanced sample classification with different backgrounds.

If the program is helpful to you, you are welcome to cite the paper. https://doi.org/10.1093/mnras/stac619

If you have any doubts. Welcome to contact me: [e-mail]

tf 1.13/1.16