https://github.com/rosinality/denoising-diffusion-pytorch
Implementation of Denoising Diffusion Probabilistic Models in PyTorch
https://github.com/rosinality/denoising-diffusion-pytorch
denoising-diffusion
Last synced: about 1 year ago
JSON representation
Implementation of Denoising Diffusion Probabilistic Models in PyTorch
- Host: GitHub
- URL: https://github.com/rosinality/denoising-diffusion-pytorch
- Owner: rosinality
- License: mit
- Created: 2020-06-24T01:21:46.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2022-06-14T10:02:07.000Z (almost 4 years ago)
- Last Synced: 2025-04-28T13:08:44.105Z (about 1 year ago)
- Topics: denoising-diffusion
- Language: Python
- Homepage:
- Size: 5.09 MB
- Stars: 379
- Watchers: 4
- Forks: 45
- Open Issues: 12
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# denoising-diffusion-pytorch
Implementation of Denoising Diffusion Probabilistic Models in PyTorch
## Installation
First please install `tensorfn`
```bash
pip install tensorfn
```
It is simple convenience library for machine learning experiments. Sorry for the inconvenience.
## Training
First prepare lmdb dataset:
```bash
python prepare_data.py --size [SIZES, e.g. 128,256] --out [LMDB NAME] [DATASET PATH]
```
Then run training looop!
```bash
python train.py --n_gpu [NUMBER OF GPUS FOR TRAINING] --conf config/diffusion.conf
```
Also, you can tried to use improved configurations from https://openreview.net/forum?id=-NEXDKk8gZ (Cosine beta schedule, more multi-headed self attention, affine modulation for time conditions, new training hyperparameters.)
```bash
python train.py --n_gpu [NUMBER OF GPUS FOR TRAINING] --conf config/improved.conf
```
## Samples
Samples from FFHQ
