Official implementation of the paper "Estimation of continuous valence and arousal levels from faces in naturalistic conditions", Antoine Toisoul, Jean Kossaifi, Adrian Bulat, Georgios Tzimiropoulos and Maja Pantic, Nature Machine Intelligence, 2021
Hello author, I noticed that when using the weights of emonet_8.pth, the test accuracy of the affectnet dataset is 38%, which significantly deviates from the ACC=0.75. Could you please let me know if there is any missing operation?
Hello author, I noticed that when using the weights of emonet_8.pth, the test accuracy of the affectnet dataset is 38%, which significantly deviates from the ACC=0.75. Could you please let me know if there is any missing operation?