Question on continuous training

[florakth]

Hi, Thanks for the nice project, I just begin with this model of pix2pixHD and would like to continuously train the second dataset based on the result model of the first dataset. I would like to know how to set the parameters like '--continue_train' and '--load_pretrain', Is there any example to set these parameters? Thank you very much !

[syfbme]

--netG local --load_pretrain the path you save for the first G There is no need to set continue_train You can refer to this script

[watertianyi]

@syfbme cuda memory exceeded？ RuntimeError: CUDA out of memory. Tried to allocate 98.00 MiB (GPU 0; 11.76 GiB total capacity; 8.86 GiB already allocated; 113.56 MiB free; 8.91 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF