Open ares123-CN opened 1 year ago
24G I guess, since I run a RTX TITAN 24G with bsz=1, it is full. OOM occurs after setting bsz>=2. By the way, I notice in the paper the authors use V100 (32G as is known to all) for training, but the config states that the bsz is 4. It puzzels me.
24G I guess, since I run a RTX TITAN 24G with bsz=1, it is full. OOM occurs after setting bsz>=2. By the way, I notice in the paper the authors use V100 (32G as is known to all) for training, but the config states that the bsz is 4. It puzzels me.
你好,请问你用24G 的GPU训练的时候有调整其他模型参数吗?我训练的时候32G都不够用,batch size已经是1了
Amazing job, I wonder the minimum GPU VRAM for training, image size is 512 and batch size is one or two