sjzhao commited on
Commit
b9b2977
1 Parent(s): ee4d9a4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +33 -14
README.md CHANGED
@@ -15,38 +15,57 @@ license: llama2
15
  - NVIDIA GPU + [CUDA](https://developer.nvidia.com/cuda-downloads)
16
 
17
  ### Installation
18
- 1. Clone repo
19
 
20
- ```bash
21
- git clone https://github.com/AILab-CVC/SEED.git
22
- cd SEED
23
- ```
 
24
 
25
- 2. Install dependent packages
26
-
27
- ```bash
28
- pip install -r requirements.txt
29
- ```
30
 
31
  ### Model Weights
32
- We provide the pretrained SEED Tokenizer and De-Tokenizer, instruction tuned SEED-LLaMA-8B and SEED-LLaMA-14B.
33
  Please download the checkpoints and save under the folder `./pretrained`.
34
 
 
 
 
 
 
 
 
35
  To reconstruct the image from the SEED visual codes using unCLIP SD-UNet, please download the pretrained [unCLIP SD](https://huggingface.co/stabilityai/stable-diffusion-2-1-unclip).
36
  Rename the checkpoint directory to **"diffusion_model"** and create a soft link to the "pretrained/seed_tokenizer" directory.
37
 
 
 
 
 
 
 
 
38
 
39
  ### Inference for visual tokenization and de-tokenization
40
  To discretize an image to 1D visual codes with causal dependency, and reconstruct the image from the visual codes using the off-the-shelf unCLIP SD-UNet:
 
41
  ```bash
 
42
  python scripts/seed_tokenizer_inference.py
43
  ```
44
 
45
- ### Launching Demo of SEED-LLaMA Locally
 
 
46
  ```bash
47
- sh start_backend.sh
48
- sh start_frontend.sh
 
 
 
49
  ```
 
 
50
 
51
  ## Citation
52
  If you find the work helpful, please consider citing:
 
15
  - NVIDIA GPU + [CUDA](https://developer.nvidia.com/cuda-downloads)
16
 
17
  ### Installation
18
+ Clone the repo and install dependent packages
19
 
20
+ ```bash
21
+ git clone https://github.com/AILab-CVC/SEED.git
22
+ cd SEED
23
+ pip install -r requirements.txt
24
+ ```
25
 
 
 
 
 
 
26
 
27
  ### Model Weights
28
+ We release the pretrained SEED Tokenizer and De-Tokenizer, instruction tuned SEED-LLaMA-8B and SEED-LLaMA-14B in [SEED Hugging Face](https://huggingface.co/AILab-CVC/SEED).
29
  Please download the checkpoints and save under the folder `./pretrained`.
30
 
31
+ ```bash
32
+ cd pretrained # SEED/pretrained
33
+ git lfs install
34
+ git clone https://huggingface.co/AILab-CVC/SEED
35
+ mv SEED/* ./
36
+ ```
37
+
38
  To reconstruct the image from the SEED visual codes using unCLIP SD-UNet, please download the pretrained [unCLIP SD](https://huggingface.co/stabilityai/stable-diffusion-2-1-unclip).
39
  Rename the checkpoint directory to **"diffusion_model"** and create a soft link to the "pretrained/seed_tokenizer" directory.
40
 
41
+ ```bash
42
+ # SEED/pretrained
43
+ git lfs install
44
+ git clone https://huggingface.co/stabilityai/stable-diffusion-2-1-unclip
45
+ mv stable-diffusion-2-1-unclip seed_tokenizer/diffusion_model
46
+ ```
47
+
48
 
49
  ### Inference for visual tokenization and de-tokenization
50
  To discretize an image to 1D visual codes with causal dependency, and reconstruct the image from the visual codes using the off-the-shelf unCLIP SD-UNet:
51
+
52
  ```bash
53
+ cd .. # SEED/
54
  python scripts/seed_tokenizer_inference.py
55
  ```
56
 
57
+ ### Launching Gradio Demo of SEED-LLaMA-14B Locally
58
+ Building the local demo of SEED-LLaMA-14B currently requires 2*32GB devices.
59
+
60
  ```bash
61
+ # SEED/
62
+ # in first terminal
63
+ sh scripts/start_backend.sh
64
+ # in second terminal
65
+ sh scripts/start_frontend.sh
66
  ```
67
+ Then the demo can be accessed through http://127.0.0.1:80
68
+
69
 
70
  ## Citation
71
  If you find the work helpful, please consider citing: