├── .gitignore ├── LICENSE ├── OUTPUT └── pretrained_model │ └── taming_dvae │ ├── taming_f16_4096_coco.yaml │ └── taming_f16_4096_ffhq.yaml ├── README.md ├── configs ├── coco_base_train.yaml ├── coco_finetune.yaml ├── ffhq_base_train.yaml └── ffhq_finetune.yaml ├── demo.ipynb ├── figures ├── clipvqdiffusion2.jpg ├── clipvqdiffusion_ffhq_page-0001.jpg └── coco_more_examples_last_page-0001.jpg ├── help_folder └── readme.md ├── image_synthesis ├── data │ ├── build.py │ ├── clip2latent_text.py │ ├── cub200_dataset.py │ ├── ffhq_dataset.py │ ├── imagenet_class_index.json │ ├── imagenet_dataset.py │ ├── mscoco_dataset.py │ └── utils │ │ ├── clip2latent_text.py │ │ ├── comm.py │ │ ├── image_preprocessor.py │ │ └── manage.py ├── distributed │ ├── distributed.py │ └── launch.py ├── engine │ ├── clip_grad_norm.py │ ├── ema.py │ ├── logger.py │ ├── lr_scheduler.py │ └── solver.py ├── modeling │ ├── build.py │ ├── codecs │ │ ├── base_codec.py │ │ ├── image_codec │ │ │ ├── clip_emb_codec.py │ │ │ ├── ema_vqvae.py │ │ │ ├── patch_vqgan.py │ │ │ └── taming_gumbel_vqvae.py │ │ └── text_codec │ │ │ └── tokenize.py │ ├── embeddings │ │ ├── base_embedding.py │ │ ├── class_embedding.py │ │ ├── clip_text_embedding.py │ │ └── dalle_mask_image_embedding.py │ ├── models │ │ ├── clip_vqdiffusion.py │ │ ├── conditional_dalle.py │ │ ├── dalle.py │ │ └── unconditional_dalle.py │ ├── modules │ │ └── clip │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── bpe_simple_vocab_16e6.txt.gz │ │ │ ├── clip.py │ │ │ ├── clip_tokenizer.py │ │ │ ├── model.py │ │ │ └── simple_tokenizer.py │ ├── transformers │ │ ├── diffusion_transformer.py │ │ └── transformer_utils.py │ └── utils │ │ └── misc.py ├── taming │ ├── __init__.py │ ├── lr_scheduler.py │ ├── models │ │ ├── cond_transformer.py │ │ └── vqgan.py │ ├── modules │ │ ├── diffusionmodules │ │ │ └── model.py │ │ ├── discriminator │ │ │ └── model.py │ │ ├── losses │ │ │ ├── __init__.py │ │ │ ├── lpips.py │ │ │ ├── segmentation.py │ │ │ └── vqperceptual.py │ │ ├── misc │ │ │ └── coord.py │ │ ├── transformer │ │ │ ├── mingpt.py │ │ │ └── permuter.py │ │ ├── util.py │ │ └── vqvae │ │ │ └── quantize.py │ └── util.py └── utils │ ├── io.py │ └── misc.py ├── install_req.sh ├── requirements.txt ├── run_test_tune.py └── train.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/LICENSE -------------------------------------------------------------------------------- /OUTPUT/pretrained_model/taming_dvae/taming_f16_4096_coco.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/OUTPUT/pretrained_model/taming_dvae/taming_f16_4096_coco.yaml -------------------------------------------------------------------------------- /OUTPUT/pretrained_model/taming_dvae/taming_f16_4096_ffhq.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/OUTPUT/pretrained_model/taming_dvae/taming_f16_4096_ffhq.yaml -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/README.md -------------------------------------------------------------------------------- /configs/coco_base_train.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/configs/coco_base_train.yaml -------------------------------------------------------------------------------- /configs/coco_finetune.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/configs/coco_finetune.yaml -------------------------------------------------------------------------------- /configs/ffhq_base_train.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/configs/ffhq_base_train.yaml -------------------------------------------------------------------------------- /configs/ffhq_finetune.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/configs/ffhq_finetune.yaml -------------------------------------------------------------------------------- /demo.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/demo.ipynb -------------------------------------------------------------------------------- /figures/clipvqdiffusion2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/figures/clipvqdiffusion2.jpg -------------------------------------------------------------------------------- /figures/clipvqdiffusion_ffhq_page-0001.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/figures/clipvqdiffusion_ffhq_page-0001.jpg -------------------------------------------------------------------------------- /figures/coco_more_examples_last_page-0001.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/figures/coco_more_examples_last_page-0001.jpg -------------------------------------------------------------------------------- /help_folder/readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/help_folder/readme.md -------------------------------------------------------------------------------- /image_synthesis/data/build.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/build.py -------------------------------------------------------------------------------- /image_synthesis/data/clip2latent_text.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/clip2latent_text.py -------------------------------------------------------------------------------- /image_synthesis/data/cub200_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/cub200_dataset.py -------------------------------------------------------------------------------- /image_synthesis/data/ffhq_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/ffhq_dataset.py -------------------------------------------------------------------------------- /image_synthesis/data/imagenet_class_index.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/imagenet_class_index.json -------------------------------------------------------------------------------- /image_synthesis/data/imagenet_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/imagenet_dataset.py -------------------------------------------------------------------------------- /image_synthesis/data/mscoco_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/mscoco_dataset.py -------------------------------------------------------------------------------- /image_synthesis/data/utils/clip2latent_text.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/utils/clip2latent_text.py -------------------------------------------------------------------------------- /image_synthesis/data/utils/comm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/utils/comm.py -------------------------------------------------------------------------------- /image_synthesis/data/utils/image_preprocessor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/utils/image_preprocessor.py -------------------------------------------------------------------------------- /image_synthesis/data/utils/manage.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/data/utils/manage.py -------------------------------------------------------------------------------- /image_synthesis/distributed/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/distributed/distributed.py -------------------------------------------------------------------------------- /image_synthesis/distributed/launch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/distributed/launch.py -------------------------------------------------------------------------------- /image_synthesis/engine/clip_grad_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/engine/clip_grad_norm.py -------------------------------------------------------------------------------- /image_synthesis/engine/ema.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/engine/ema.py -------------------------------------------------------------------------------- /image_synthesis/engine/logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/engine/logger.py -------------------------------------------------------------------------------- /image_synthesis/engine/lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/engine/lr_scheduler.py -------------------------------------------------------------------------------- /image_synthesis/engine/solver.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/engine/solver.py -------------------------------------------------------------------------------- /image_synthesis/modeling/build.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/build.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/base_codec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/base_codec.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/image_codec/clip_emb_codec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/image_codec/clip_emb_codec.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/image_codec/ema_vqvae.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/image_codec/ema_vqvae.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/image_codec/patch_vqgan.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/image_codec/patch_vqgan.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/image_codec/taming_gumbel_vqvae.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/image_codec/taming_gumbel_vqvae.py -------------------------------------------------------------------------------- /image_synthesis/modeling/codecs/text_codec/tokenize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/codecs/text_codec/tokenize.py -------------------------------------------------------------------------------- /image_synthesis/modeling/embeddings/base_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/embeddings/base_embedding.py -------------------------------------------------------------------------------- /image_synthesis/modeling/embeddings/class_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/embeddings/class_embedding.py -------------------------------------------------------------------------------- /image_synthesis/modeling/embeddings/clip_text_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/embeddings/clip_text_embedding.py -------------------------------------------------------------------------------- /image_synthesis/modeling/embeddings/dalle_mask_image_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/embeddings/dalle_mask_image_embedding.py -------------------------------------------------------------------------------- /image_synthesis/modeling/models/clip_vqdiffusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/models/clip_vqdiffusion.py -------------------------------------------------------------------------------- /image_synthesis/modeling/models/conditional_dalle.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/models/conditional_dalle.py -------------------------------------------------------------------------------- /image_synthesis/modeling/models/dalle.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/models/dalle.py -------------------------------------------------------------------------------- /image_synthesis/modeling/models/unconditional_dalle.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/models/unconditional_dalle.py -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/README.md: -------------------------------------------------------------------------------- 1 | https://github.com/openai/CLIP -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/__init__.py: -------------------------------------------------------------------------------- 1 | from .clip import * 2 | -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/bpe_simple_vocab_16e6.txt.gz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/modules/clip/bpe_simple_vocab_16e6.txt.gz -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/clip.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/modules/clip/clip.py -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/clip_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/modules/clip/clip_tokenizer.py -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/modules/clip/model.py -------------------------------------------------------------------------------- /image_synthesis/modeling/modules/clip/simple_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/modules/clip/simple_tokenizer.py -------------------------------------------------------------------------------- /image_synthesis/modeling/transformers/diffusion_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/transformers/diffusion_transformer.py -------------------------------------------------------------------------------- /image_synthesis/modeling/transformers/transformer_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/transformers/transformer_utils.py -------------------------------------------------------------------------------- /image_synthesis/modeling/utils/misc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/modeling/utils/misc.py -------------------------------------------------------------------------------- /image_synthesis/taming/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /image_synthesis/taming/lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/lr_scheduler.py -------------------------------------------------------------------------------- /image_synthesis/taming/models/cond_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/models/cond_transformer.py -------------------------------------------------------------------------------- /image_synthesis/taming/models/vqgan.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/models/vqgan.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/diffusionmodules/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/diffusionmodules/model.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/discriminator/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/discriminator/model.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/losses/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/losses/__init__.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/losses/lpips.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/losses/lpips.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/losses/segmentation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/losses/segmentation.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/losses/vqperceptual.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/losses/vqperceptual.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/misc/coord.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/misc/coord.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/transformer/mingpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/transformer/mingpt.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/transformer/permuter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/transformer/permuter.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/util.py -------------------------------------------------------------------------------- /image_synthesis/taming/modules/vqvae/quantize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/modules/vqvae/quantize.py -------------------------------------------------------------------------------- /image_synthesis/taming/util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/taming/util.py -------------------------------------------------------------------------------- /image_synthesis/utils/io.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/utils/io.py -------------------------------------------------------------------------------- /image_synthesis/utils/misc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/image_synthesis/utils/misc.py -------------------------------------------------------------------------------- /install_req.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/install_req.sh -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/requirements.txt -------------------------------------------------------------------------------- /run_test_tune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/run_test_tune.py -------------------------------------------------------------------------------- /train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/INFINIQ-AI1/CLIPVQDiffusion/HEAD/train.py --------------------------------------------------------------------------------