code/train.py @@ -251,7 +251,7 @@ loss_pitch += alpha * loss_reg_pitch loss_roll += alpha * loss_reg_roll loss_yaw *= 0.35 loss_yaw *= 1 # Finetuning loss loss_seq = [loss_yaw, loss_pitch, loss_roll]
