upload weights

README.md

@@ -0,0 +1,116 @@
+---
+license: creativeml-openrail-m
+tags:
+- text-to-image
+---
+
+# Karlo v1 alpha
+
+Karlo is a text-conditional image generation model based on OpenAI's unCLIP architecture with the improvement over the standard super-resolution model from 64px to 256px, recovering high-frequency details only in the small number of denoising steps.
+
+* [Original codebase](https://github.com/kakaobrain/karlo)
+
+## Usage
+
+Karlo is available in diffusers!
+
+```python
+pip install diffusers transformers accelerate safetensors
+```
+
+### Text to image
+
+```python
+from diffusers import UnCLIPPipeline
+import torch
+
+pipe = UnCLIPPipeline.from_pretrained("kakaobrain/karlo-v1-alpha", torch_dtype=torch.float16)
+pipe = pipe.to('cuda')
+
+prompt = "a high-resolution photograph of a big red frog on a green leaf."
+
+image = pipe([prompt]).images[0]
+
+image.save("./frog.png")
+```
+
+![img](https://huggingface.co/datasets/patrickvonplaten/images/resolve/main/frog.png)
+
+### Image variation
+
+```python
+from diffusers import UnCLIPImageVariationPipeline
+import torch
+from PIL import Image
+
+pipe = UnCLIPImageVariationPipeline.from_pretrained("kakaobrain/karlo-v1-alpha-image-variations", torch_dtype=torch.float16)
+pipe = pipe.to('cuda')
+
+image = Image.open("./frog.png")
+
+image = pipe(image).images[0]
+
+image.save("./frog-variation.png")
+```
+
+![img](https://huggingface.co/datasets/williamberman/images/resolve/main/frog-variation.png)
+
+## Model Architecture
+
+### Overview
+Karlo is a text-conditional diffusion model based on unCLIP, composed of prior, decoder, and super-resolution modules. In this repository, we include the improved version of the standard super-resolution module for upscaling 64px to 256px only in 7 reverse steps, as illustrated in the figure below:
+
+<p float="left">
+  <img src="https://raw.githubusercontent.com/kakaobrain/karlo/main/assets/improved_sr_arch.jpg"/>
+</p>
+
+In specific, the standard SR module trained by DDPM objective upscales 64px to 256px in the first 6 denoising steps based on the respacing technique. Then, the additional fine-tuned SR module trained by [VQ-GAN](https://compvis.github.io/taming-transformers/)-style loss performs the final reverse step to recover high-frequency details. We observe that this approach is very effective to upscale the low-resolution in a small number of reverse steps.
+
+### Details
+We train all components from scratch on 115M image-text pairs including COYO-100M, CC3M, and CC12M. In the case of Prior and Decoder, we use ViT-L/14 provided by OpenAI’s [CLIP repository](https://github.com/openai/CLIP). Unlike the original implementation of unCLIP, we replace the trainable transformer in the decoder into the text encoder in ViT-L/14 for efficiency. In the case of the SR module, we first train the model using the DDPM objective in 1M steps, followed by additional 234K steps to fine-tune the additional component. The table below summarizes the important statistics of our components:
+
+| | Prior | Decoder | SR |
+|:------|----:|----:|----:|
+| CLIP | ViT-L/14 | ViT-L/14 | - |
+| #param | 1B | 900M | 700M + 700M |
+| #optimization steps | 1M | 1M | 1M + 0.2M |
+| #sampling steps | 25 | 50 (default), 25 (fast) | 7 |
+|Checkpoint links| [ViT-L-14](https://arena.kakaocdn.net/brainrepo/models/karlo-public/v1.0.0.alpha/096db1af569b284eb76b3881534822d9/ViT-L-14.pt), [ViT-L-14 stats](https://arena.kakaocdn.net/brainrepo/models/karlo-public/v1.0.0.alpha/0b62380a75e56f073e2844ab5199153d/ViT-L-14_stats.th), [model](https://arena.kakaocdn.net/brainrepo/models/karlo-public/v1.0.0.alpha/efdf6206d8ed593961593dc029a8affa/decoder-ckpt-step%3D01000000-of-01000000.ckpt) | [model](https://arena.kakaocdn.net/brainrepo/models/karlo-public/v1.0.0.alpha/85626483eaca9f581e2a78d31ff905ca/prior-ckpt-step%3D01000000-of-01000000.ckpt) | [model](https://arena.kakaocdn.net/brainrepo/models/karlo-public/v1.0.0.alpha/4226b831ae0279020d134281f3c31590/improved-sr-ckpt-step%3D1.2M.ckpt) |
+
+In the checkpoint links, ViT-L-14 is equivalent to the original version, but we include it for convenience. We also remark that ViT-L-14-stats is required to normalize the outputs of the prior module.
+
+### Evaluation
+We quantitatively measure the performance of Karlo-v1.0.alpha in the validation split of CC3M and MS-COCO. The table below presents CLIP-score and FID. To measure FID, we resize the image of the shorter side to 256px, followed by cropping it at the center. We set classifier-free guidance scales for prior and decoder to 4 and 8 in all cases. We observe that our model achieves reasonable performance even with 25 sampling steps of decoder. 
+
+CC3M
+| Sampling step | CLIP-s (ViT-B/16) | FID (13k from val)|
+|:------|----:|----:|
+| Prior (25) + Decoder (25) + SR (7) | 0.3081 | 14.37 |
+| Prior (25) + Decoder (50) + SR (7) | 0.3086 | 13.95 |
+
+MS-COCO
+| Sampling step | CLIP-s (ViT-B/16) | FID (30k from val)|
+|:------|----:|----:|
+| Prior (25) + Decoder (25) + SR (7) | 0.3192 | 15.24 |
+| Prior (25) + Decoder (50) + SR (7) | 0.3192 | 14.43 |
+
+
+For more information, please refer to the upcoming technical report.
+
+### Training Details
+
+This alpha version of Karlo is trained on 115M image-text pairs, 
+including [COYO](https://github.com/kakaobrain/coyo-dataset)-100M high-quality subset, CC3M, and CC12M. 
+For those who are interested in a better version of Karlo trained on more large-scale high-quality datasets, 
+please visit the landing page of our application [B^DISCOVER](https://bdiscover.kakaobrain.com/).
+
+## BibTex
+If you find this repository useful in your research, please cite:
+```
+@misc{kakaobrain2022karlo-v1-alpha,
+  title         = {Karlo-v1.0.alpha on COYO-100M and CC15M},
+  author        = {Donghoon Lee, Jiseob Kim, Jisu Choi, Jongmin Kim, Minwoo Byeon, Woonhyuk Baek and Saehoon Kim},
+  year          = {2022},
+  howpublished  = {\url{https://github.com/kakaobrain/karlo}},
+}
+```

decoder/config.json

@@ -0,0 +1,45 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.12.0.dev0",
+  "_name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/decoder",
+  "act_fn": "silu",
+  "attention_head_dim": 64,
+  "block_out_channels": [
+    320,
+    640,
+    960,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": "identity",
+  "cross_attention_dim": 1536,
+  "down_block_types": [
+    "ResnetDownsampleBlock2D",
+    "SimpleCrossAttnDownBlock2D",
+    "SimpleCrossAttnDownBlock2D",
+    "SimpleCrossAttnDownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dual_cross_attention": false,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 3,
+  "layers_per_block": 3,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DSimpleCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 6,
+  "resnet_time_scale_shift": "scale_shift",
+  "sample_size": 64,
+  "up_block_types": [
+    "SimpleCrossAttnUpBlock2D",
+    "SimpleCrossAttnUpBlock2D",
+    "SimpleCrossAttnUpBlock2D",
+    "ResnetUpsampleBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": false
+}

decoder/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df4c1b9bceeceb92a88ff252e964be3a39308e799fdf01d5557d247f11868310
+size 3542027081

decoder_scheduler/scheduler_config.json

@@ -0,0 +1,9 @@
+{
+  "_class_name": "UnCLIPScheduler",
+  "_diffusers_version": "0.12.0.dev0",
+  "clip_sample": true,
+  "clip_sample_range": 1.0,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "variance_type": "learned_range"
+}

feature_extractor/preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 224
+  }
+}

image_encoder/config.json

@@ -0,0 +1,23 @@
+{
+  "_name_or_path": "openai/clip-vit-large-patch14",
+  "architectures": [
+    "CLIPVisionModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "dropout": 0.0,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 1024,
+  "image_size": 224,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "model_type": "clip_vision_model",
+  "num_attention_heads": 16,
+  "num_channels": 3,
+  "num_hidden_layers": 24,
+  "patch_size": 14,
+  "projection_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1"
+}

image_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c304e9a1c99835f6c2d355cbb5438e7e6204d4903b2f1b056120be4bd5050f1f
+size 1215996977

model_index.json

@@ -0,0 +1,44 @@
+{
+  "_class_name": "UnCLIPImageVariationPipeline",
+  "_diffusers_version": "0.12.0.dev0",
+  "decoder": [
+    "diffusers",
+    "UNet2DConditionModel"
+  ],
+  "decoder_scheduler": [
+    "diffusers",
+    "UnCLIPScheduler"
+  ],
+  "feature_extractor": [
+    "transformers",
+    "CLIPImageProcessor"
+  ],
+  "image_encoder": [
+    "transformers",
+    "CLIPVisionModelWithProjection"
+  ],
+  "super_res_first": [
+    "diffusers",
+    "UNet2DModel"
+  ],
+  "super_res_last": [
+    "diffusers",
+    "UNet2DModel"
+  ],
+  "super_res_scheduler": [
+    "diffusers",
+    "UnCLIPScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "CLIPTextModelWithProjection"
+  ],
+  "text_proj": [
+    "unclip",
+    "UnCLIPTextProjModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "CLIPTokenizer"
+  ]
+}

super_res_first/config.json

@@ -0,0 +1,39 @@
+{
+  "_class_name": "UNet2DModel",
+  "_diffusers_version": "0.12.0.dev0",
+  "_name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/super_res_first",
+  "act_fn": "silu",
+  "add_attention": false,
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    960,
+    1280
+  ],
+  "center_input_sample": false,
+  "down_block_types": [
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D"
+  ],
+  "downsample_padding": 1,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 6,
+  "layers_per_block": 3,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 256,
+  "time_embedding_type": "positional",
+  "up_block_types": [
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D"
+  ]
+}

super_res_first/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be101081e8dba0d7dfe19845187c8efe6d4c68e32daa33224d1f8631333085b3
+size 2766471473

super_res_last/config.json

@@ -0,0 +1,39 @@
+{
+  "_class_name": "UNet2DModel",
+  "_diffusers_version": "0.12.0.dev0",
+  "_name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/super_res_last",
+  "act_fn": "silu",
+  "add_attention": false,
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    960,
+    1280
+  ],
+  "center_input_sample": false,
+  "down_block_types": [
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D",
+    "ResnetDownsampleBlock2D"
+  ],
+  "downsample_padding": 1,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 6,
+  "layers_per_block": 3,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "resnet_time_scale_shift": "default",
+  "sample_size": null,
+  "time_embedding_type": "positional",
+  "up_block_types": [
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D",
+    "ResnetUpsampleBlock2D"
+  ]
+}

super_res_last/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3356695fc89b8fa39b3f1ec1d3d635cbc065387ef622a04593897aaae7247f2
+size 2766471473

super_res_scheduler/scheduler_config.json

@@ -0,0 +1,9 @@
+{
+  "_class_name": "UnCLIPScheduler",
+  "_diffusers_version": "0.12.0.dev0",
+  "clip_sample": true,
+  "clip_sample_range": 1.0,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "variance_type": "fixed_small_log"
+}

text_encoder/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/text_encoder",
+  "architectures": [
+    "CLIPTextModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 768,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "projection_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1",
+  "vocab_size": 49408
+}

text_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aec40a3f65f96219cb30a80554a39fcb365146b6aa7d0cb9621e08b010ba672d
+size 494666597

text_proj/config.json

@@ -0,0 +1,9 @@
+{
+  "_class_name": "UnCLIPTextProjModel",
+  "_diffusers_version": "0.12.0.dev0",
+  "_name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/text_proj",
+  "clip_embeddings_dim": 768,
+  "clip_extra_context_tokens": 4,
+  "cross_attention_dim": 1536,
+  "time_embed_dim": 1280
+}

text_proj/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46c3cb5d013e22221d156a9f30ef9221a4f9cfc547ab295cff870b2e26405a62
+size 31517844

tokenizer/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "!",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,35 @@
+{
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "device_map": "auto",
+  "do_lower_case": true,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 77,
+  "name_or_path": "/home/william_huggingface_co/.cache/huggingface/diffusers/models--kakaobrain--karlo-v1-alpha/snapshots/c8dc0970acf0db0b6b4efb8c47adc22072b4a67d/tokenizer",
+  "pad_token": "!",
+  "special_tokens_map_file": "./special_tokens_map.json",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}