Inpaint instal the new for apple1/23/2024 ![]() save(, "ip_adapter.bin") Third-party Usage Import torch ckpt = "checkpoint-50000/pytorch_model.bin" sd = torch. (Maybe this training strategy can also be used to speed up the training of controlnet).įor training, you should install accelerate and make your own dataset into a json file. Then, we employ a multi-scale strategy for fine-tuning. Firstly, we perform pre-training at a resolution of 512x512. However, in the new version, we have implemented a more effective two-stage training strategy. A Faster and better training recipe: In our previous version, training directly at a resolution of 1024x1024 proved to be highly inefficient.Although ViT-bigG is much larger than ViT-H, our experimental results did not find a significant difference, and the smaller model can reduce the memory usage in the inference phase. Switch to CLIP-ViT-H: we trained the new IP-Adapter with OpenCLIP-ViT-H-14 instead of OpenCLIP-ViT-bigG-14.The comparison of IP-Adapter_XL with Reimagine XL is shown as follows: ip_adapter_sdxl_controlnet_demo: structural generation with image prompt.ip_adapter_sdxl_demo: image variations with image prompt.But you can just resize to 224x224 for non-square images, the comparison is as follows: For the non square images, it will miss the information outside the center. For the version of SD 1.5, we recommend using community models to generate good images.Īs the image is center cropped in the default image processor of CLIP, IP-Adapter works best for square images. In most cases, setting scale=0.5 can get good results. For multimodal prompts, you can adjust the scale to get the best results.If you lower the scale, more diverse images can be generated, but they may not be as consistent with the image prompt. "best quality", you can also use any negative text prompt). If you only use the image prompt, you can set the scale=1.0 and text_prompt=""(or some generic text prompts, e.g.ip_adapter-plus-face_demo: generation with face image as prompt.ip_adapter-plus_demo: the demo of IP-Adapter with fine-grained features.ip_adapter_multimodal_prompts_demo: generation with multimodal prompts.ip_adapter_controlnet_demo, ip_adapter_t2i-adapter: structural generation with image prompt. ![]()
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |