hello, when i train the model on multiple gpu cards, the loss calculation always cause errors for the misalignment in dimension 0, i'm awared of the batchsize have changed for different gpus, but i don't know what to deal with. does the code can train on multiple gpu cards ? if can, what should i change to fix the problem. looking forward to your reply.
Besides, i want to ask you if you can share the configuration file and dataset file on 3DHP , I want to train in this dataset and i'm not very clear about the configuration about the dataset, thank you!
hello, when i train the model on multiple gpu cards, the loss calculation always cause errors for the misalignment in dimension 0, i'm awared of the batchsize have changed for different gpus, but i don't know what to deal with. does the code can train on multiple gpu cards ? if can, what should i change to fix the problem. looking forward to your reply.
Besides, i want to ask you if you can share the configuration file and dataset file on 3DHP , I want to train in this dataset and i'm not very clear about the configuration about the dataset, thank you!