Logs when coding
如何从from_pretrained中加载safetensors：先下载hugging face的model card到本地，然后将名字修改为默认的比如“diffusion_model.safetensors”，safetensors参考，safetensors参考2
运行Lora时，train_text_to_image_lora.py raise ValueError(“Attempting to unscale FP16 gradients.”) ，解决报错问题：不从命令行传递mixed_presision参数，直接改用默认参数：参考链接
PIL.Image格式的图片在调用resize方法的时候,高宽参数的输入顺序是w，h而不是h,w
This post is licensed under CC BY 4.0 by the author.