CNGaoWenbo

Results 2 comments of CNGaoWenbo

I only changed dist_util `GPUS_PER_NODE = 4 #change to 4 SETUP_RETRY_COUNT = 3 def setup_dist(): if dist.is_initialized(): return os.environ["CUDA_VISIBLE_DEVICES"] = '6,7,8,9' #change to '6,7,8,9' backend = "gloo" if not th.cuda.is_available()...

thank you. From my understanding, the pre-trained seg_mask_unet outputs a generated picture at t, then feed this picture into the dynamic diffuser(seg_mask_confidence_predictor in codes). I understand that seg_mask_unet is trained,...