 This paper presents a novel approach to classifying satellite image time series, SITS, using temporal convolutional neural networks, Temp-CNNs. Compared to existing methods, Temp-CNNs were found to be more accurate and efficient for SITS classification. Additionally, the authors provided guidelines for network architecture, regularization mechanisms, and hyperparameters. Furthermore, they assessed the visual quality of the land cover maps produced by Temp-CNNs.