tensorflow_Realtime_Multi-Person_Pose_Estimation
tensorflow_Realtime_Multi-Person_Pose_Estimation copied to clipboard
trained 60 epoch,the validation loss of paf and heatmap is still decrease
hello,I pull down your code almost no change.I trained the mobilenet based model on 2080Ti with 60 epoch.and I found the val_loss is still decrease
the x-axis is the epoch from 1 to 60,the y-axis is loss.
That's good. The lower val loss, the better.
Excuse me,Are you using the same learning strategy as train_custom_loop_mobilenet.py?