Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,36 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
4 |
-
anywefadoor
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
4 |
+
anywefadoor
|
5 |
+
|
6 |
+
## Installation
|
7 |
+
|
8 |
+
various necesary libs to install via apt will reveal themselves depending on hardware. libxext and others after the conda install and run first time.
|
9 |
+
|
10 |
+
Install with `conda`:
|
11 |
+
```bash
|
12 |
+
conda env create -f environment.yaml
|
13 |
+
conda activate anydoor
|
14 |
+
```
|
15 |
+
or `pip`:
|
16 |
+
```bash
|
17 |
+
pip install -r requirements.txt
|
18 |
+
```
|
19 |
+
Additionally, for training, you need to install panopticapi, pycocotools, and lvis-api.
|
20 |
+
```bash
|
21 |
+
pip install git+https://github.com/cocodataset/panopticapi.git
|
22 |
+
|
23 |
+
pip install pycocotools -i https://pypi.douban.com/simple
|
24 |
+
|
25 |
+
pip install lvis
|
26 |
+
```
|
27 |
+
## Download Checkpoints
|
28 |
+
Download WefaDoor checkpoint:
|
29 |
+
* [HuggingFace](https://huggingface.co/thisisAce/ADOOR_ACE/blob/main/models/adbase-step%3D41250.ckpt)
|
30 |
+
|
31 |
+
|
32 |
+
Download DINOv2 checkpoint and revise `/configs/anydoor.yaml` for the path (line 83)
|
33 |
+
* URL: https://github.com/facebookresearch/dinov2?tab=readme-ov-file
|
34 |
+
|
35 |
+
Download Stable Diffusion V2.1 if you want to train from scratch.
|
36 |
+
* URL: https://huggingface.co/thisisAce/ADOOR_ACE/blob/main/models/dinov2_vitg14_pretrain.pth
|