 We developed a deep learning model to predict whether or not a patient will respond to repetitive transcranial magnetic stimulation, RTMS. The model used time-frequency representations of electroencephalography, EEG, data from the frontal lobe of the brain and convolutional recurrent neural networks, CRNNs, equipped with an attention mechanism. This combination of techniques achieved high accuracy in predicting the response to RTMS treatment. Furthermore, we tested the model's generalizability on a public dataset and found it to have high accuracy as well. Thus, this model provides a promising platform for the prediction of response to RTMS treatment. This article was authored by Mohsen Sadat Shahabi, Ahmed Shalbath, Riza Rostami, and others.