FACT: Autoencoder and Attention Conv-LSTM-Based Collaborative Framework for Cloud Cover Prediction | IEEE Journals & Magazine | IEEE Xplore