Text-to-Image
Diffusers
English
xingpng commited on
Commit
7e7d9b0
·
1 Parent(s): 9d97782
Files changed (1) hide show
  1. README.md +12 -4
README.md CHANGED
@@ -15,8 +15,8 @@ license: apache-2.0
15
 
16
  <a href='https://csgo-gen.github.io/'><img src='https://img.shields.io/badge/Project-Page-green'></a>
17
  <a href='https://arxiv.org/abs/2404.02733'><img src='https://img.shields.io/badge/Technique-Report-red'></a>
18
- [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Space-red)](https://huggingface.co/spaces/InstantX/InstantStyle)
19
- [![ModelScope](https://img.shields.io/badge/ModelScope-Studios-blue)](https://modelscope.cn/studios/instantx/InstantStyle/summary)
20
 
21
  </div>
22
 
@@ -55,6 +55,15 @@ license: apache-2.0
55
  This repo, named **CSGO**, contains the official PyTorch implementation of our paper [CSGO: Content-Style Composition in Text-to-Image Generation](https://arxiv.org/pdf/).
56
  We are actively updating and improving this repository. If you find any bugs or have suggestions, welcome to raise issues or submit pull requests (PR) 💖.
57
 
 
 
 
 
 
 
 
 
 
58
  ## Pipeline 💻
59
  <p align="center">
60
  <img src="assets/image3_1.jpg">
@@ -143,12 +152,11 @@ controlnet_target_content_blocks = controlnet_BLOCKS['content']
143
  controlnet_target_style_blocks = controlnet_BLOCKS['style']
144
 
145
  csgo = CSGO(pipe, image_encoder_path, csgo_ckpt, device, num_content_tokens=4,num_style_tokens=32,
146
- target_content_blocks=target_content_blocks, target_style_blocks=target_style_blocks,controlnet=False,controlnet_adapter=True,
147
  controlnet_target_content_blocks=controlnet_target_content_blocks,
148
  controlnet_target_style_blocks=controlnet_target_style_blocks,
149
  content_model_resampler=True,
150
  style_model_resampler=True,
151
- load_controlnet=False,
152
 
153
  )
154
 
 
15
 
16
  <a href='https://csgo-gen.github.io/'><img src='https://img.shields.io/badge/Project-Page-green'></a>
17
  <a href='https://arxiv.org/abs/2404.02733'><img src='https://img.shields.io/badge/Technique-Report-red'></a>
18
+ [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Space-red)](https://huggingface.co/spaces/InstantX/CSGO)
19
+
20
 
21
  </div>
22
 
 
55
  This repo, named **CSGO**, contains the official PyTorch implementation of our paper [CSGO: Content-Style Composition in Text-to-Image Generation](https://arxiv.org/pdf/).
56
  We are actively updating and improving this repository. If you find any bugs or have suggestions, welcome to raise issues or submit pull requests (PR) 💖.
57
 
58
+ ## Detail ✨
59
+ We currently release two model weights.
60
+
61
+ | Mode | content token | style token | Other |
62
+ |:------------:|:-----------:|:-----------:|:---------------:|
63
+ |csgo.bin|4|16| - |
64
+ |csgo_4_32.bin|4|32| Deepspeed zero2 |
65
+
66
+
67
  ## Pipeline 💻
68
  <p align="center">
69
  <img src="assets/image3_1.jpg">
 
152
  controlnet_target_style_blocks = controlnet_BLOCKS['style']
153
 
154
  csgo = CSGO(pipe, image_encoder_path, csgo_ckpt, device, num_content_tokens=4,num_style_tokens=32,
155
+ target_content_blocks=target_content_blocks, target_style_blocks=target_style_blocks,controlnet_adapter=True,
156
  controlnet_target_content_blocks=controlnet_target_content_blocks,
157
  controlnet_target_style_blocks=controlnet_target_style_blocks,
158
  content_model_resampler=True,
159
  style_model_resampler=True,
 
160
 
161
  )
162