Deep Neural Networks of Solar Flare Forecasting for Complex Active Regions

There are four models of Complex Active Region Flare Forecast Model-1(CARFFM-1), Complex Active Region Flare Forecast Model-2(CARFFM-2), Complex Active Region Flare Forecast Model-3(CARFFM-3), and Complex Active Region Flare Forecast Model-4(CARFFM-4). For each model, the corresponding folder contains four subfolders of "code", "model", "train" and "apply", which are the modeling codes in Python language, the model after training, the training set, and the test set, respectively.

In "code" subfolder, there are 4 files of "my_genetateds.py", "my_forward.py", "my_backward.py", and "my_apply.py". "my_apply.py" is used to test the model.