Closed flyyyyer closed 1 year ago
Did you adjust batch_size and learning_rate in the code? Using the most recent PyTorch for the installation guide seems to be recommended.
I change the batch_size from 32 to 4 and 8, because I only have one 3090 GPU with 24GB.
In my opinion, if you change Learning_Rate from 0.0005 to 0.0001 (or lower), you will get similar results.
Hi, if you change the batch size, you may need to adjust the lr accordingly. Thank you @HoBeom for pointing that out!
I want to reproduce the result of "3D Pose (H36M-SH, scratch), 39.1mm", but I only can get 40.0mm. So I want to know what is the PyTorch version you used to train the model?