Upload folder using huggingface_hub
#1
by
pcuenq
HF staff
- opened
- .gitattributes +1 -0
- README.md +45 -0
- samples.png +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
samples.png filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apple-ascl
|
3 |
+
tags:
|
4 |
+
- mdm
|
5 |
+
---
|
6 |
+
|
7 |
+
# Matryoshka Diffusion Models
|
8 |
+
|
9 |
+
Matryoshka Diffusion Models was introduced in [the paper of the same name](https://huggingface.co/papers/2310.15111), by Jiatao Gu,Shuangfei Zhai, Yizhe Zhang, Josh Susskind, Navdeep Jaitly.
|
10 |
+
|
11 |
+
This repository contains the **Flickr 64** checkpoint.
|
12 |
+
|
13 |
+
![Generation Examples from the MDM repository](samples.png)
|
14 |
+
|
15 |
+
### Highlights
|
16 |
+
|
17 |
+
* This checkpoint was trained on a dataset of 50M text-image pairs collected from Flickr.
|
18 |
+
* This model was trained using a single UNet (not nested), and generates images with a resolution of 64 × 64.
|
19 |
+
* Despite training on relatively small datasets, MDMs show strong zero-shot capabilities of generating high-resolution images and videos.
|
20 |
+
|
21 |
+
## Checkpoints
|
22 |
+
|
23 |
+
| Model | Dataset | Resolution | Nested UNets |
|
24 |
+
|---------------------------------------------------------|------------|-------------|--------------|
|
25 |
+
| [mdm-flickr-64](https://hf.co/pcuenq/mdm-flickr-64) | Flickr 50M | 64 × 64 | ❎ |
|
26 |
+
| [mdm-flickr-256](https://hf.co/pcuenq/mdm-flickr-256) | Flickr 50M | 256 × 256 | ✅ |
|
27 |
+
| [mdm-flickr-1024](https://hf.co/pcuenq/mdm-flickr-1024) | Flickr 50M | 1024 × 1024 | ✅ |
|
28 |
+
|
29 |
+
## How to Use
|
30 |
+
|
31 |
+
Please, refer to the [original repository](https://github.com/apple/ml-mdm) for training and inference instructions.
|
32 |
+
|
33 |
+
## Citation
|
34 |
+
|
35 |
+
```
|
36 |
+
@misc{gu2023matryoshkadiffusionmodels,
|
37 |
+
title={Matryoshka Diffusion Models},
|
38 |
+
author={Jiatao Gu and Shuangfei Zhai and Yizhe Zhang and Josh Susskind and Navdeep Jaitly},
|
39 |
+
year={2023},
|
40 |
+
eprint={2310.15111},
|
41 |
+
archivePrefix={arXiv},
|
42 |
+
primaryClass={cs.CV},
|
43 |
+
url={https://arxiv.org/abs/2310.15111},
|
44 |
+
}
|
45 |
+
```
|
samples.png
ADDED
Git LFS Details
|