Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

https://github.com/DmitryUlyanov/AGE

Code for the paper "Adversarial Generator-Encoder Networks"
https://github.com/DmitryUlyanov/AGE

gan

Last synced: 2 months ago
JSON representation

Code for the paper "Adversarial Generator-Encoder Networks"

Host: GitHub
URL: https://github.com/DmitryUlyanov/AGE
Owner: DmitryUlyanov
Created: 2017-04-07T16:13:21.000Z (about 7 years ago)
Default Branch: master
Last Pushed: 2018-06-28T15:56:34.000Z (almost 6 years ago)
Last Synced: 2024-01-25T20:08:57.375Z (5 months ago)
Topics: gan
Language: Python
Homepage: https://arxiv.org/abs/1704.02304
Size: 20.8 MB
Stars: 283
Watchers: 14
Forks: 50
Open Issues: 6
Metadata Files:
- Readme: README.md

Lists

Awesome-pytorch-list - AGE - Encoder Networks" by Dmitry Ulyanov, Andrea Vedaldi and Victor Lempitsky which can be found [here](http://sites.skoltech.ru/app/data/uploads/sites/25/2017/04/AGE.pdf) (Paper implementations / Other libraries:)
Awesome-pytorch-list-CNVersion - AGE - Encoder Networks](http://sites.skoltech.ru/app/data/uploads/sites/25/2017/04/AGE.pdf)）。 (Paper implementations｜论文实现 / Other libraries｜其他库:)

README

This repository contains code for the paper

**["Adversarial Generator-Encoder Networks"](https://arxiv.org/abs/1704.02304)** (AAAI'18) by *Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky*.

![](data/readme_pics/age.png)

## Pretrained models

This is how you can access the models used to generate figures in the paper.

1) First install dev version of `pytorch 0.2` and make sure you have `jupyter notebook` ready.

2) Then download the models with the script:
```
bash download_pretrained.sh
```

3) Run `jupyter notebook` and go through `evaluate.ipynb`.

Here is an example of samples and reconstructions for `imagenet`, `celeba` and `cifar10` datasets generated with `evaluate.ipynb`.

#### Celeba

#### Cifar10

#### Tiny ImageNet

# Training

Use `age.py` script to train a model. Here are the most important parameters:

* `--dataset`: one of [celeba, cifar10, imagenet, svhn, mnist]
* `--dataroot`: for datasets included in `torchvision` it is a directory where everything will be downloaded to; for imagenet, celeba datasets it is a path to a directory with folders `train` and `val` inside.
* `--image_size`:
* `--save_dir`: path to a folder, where checkpoints will be stored
* `--nz`: dimensionality of latent space
* `-- batch_size`: Batch size. Default 64.
* `--netG`: `.py` file with generator definition. Searched in `models` directory
* `--netE`: `.py` file with generator definition. Searched in `models` directory
* `--netG_chp`: path to a generator checkpoint to load from
* `--netE_chp`: path to an encoder checkpoint to load from
* `--nepoch`: number of epoch to run
* `--start_epoch`: epoch number to start from. Useful for finetuning.
* `--e_updates`: Update plan for encoder. `;KL_fake:,KL_real:,match_z:,match_x:`.
* `--g_updates`: Update plan for generator. `;KL_fake:,match_z:,match_x:`.

And misc arguments:
* `--workers`: number of dataloader workers.
* `--ngf`: controlles number of channels in generator
* `--ndf`: controlles number of channels in encoder
* `--beta1`: parameter for ADAM optimizer
* `--cpu`: do not use GPU
* `--criterion`: Parametric `param` or non-parametric `nonparam` way to compute KL. Parametric fits Gaussian into data, non-parametric is based on nearest neighbors. Default: `param`.
* `--KL`: What KL to compute: `qp` or `pq`. Default is `qp`.
* `--noise`: `sphere` for uniform on sphere or `gaussian`. Default `sphere`.
* `--match_z`: loss to use as reconstruction loss in latent space. `L1|L2|cos`. Default `cos`.
* `--match_x`: loss to use as reconstruction loss in data space. `L1|L2|cos`. Default `L1`.
* `--drop_lr`: each `drop_lr` epochs a learning rate is dropped.
* `--save_every`: controls how often intermediate results are stored. Default `50`.
* `--manual_seed`: random seed. Default `123`.

Here is `cmd` you can start with:

### Celeba
Let `data_root` to be a directory with two folders `train`, `val`, each with the images for corresponding split.

```
python age.py --dataset celeba --dataroot --image_size 64 --save_dir --lr 0.0002 --nz 64 --batch_size 64 --netG dcgan64px --netE dcgan64px --nepoch 5 --drop_lr 5 --e_updates '1;KL_fake:1,KL_real:1,match_z:0,match_x:10' --g_updates '3;KL_fake:1,match_z:1000,match_x:0'
```

It is beneficial to finetune the model with larger `batch_size` and stronger matching weight then:
```
python age.py --dataset celeba --dataroot --image_size 64 --save_dir --start_epoch 5 --lr 0.0002 --nz 64 --batch_size 256 --netG dcgan64px --netE dcgan64px --nepoch 6 --drop_lr 5 --e_updates '1;KL_fake:1,KL_real:1,match_z:0,match_x:15' --g_updates '3;KL_fake:1,match_z:1000,match_x:0' --netE_chp /netE_epoch_5.pth --netG_chp /netG_epoch_5.pth
```

### Imagenet

```
python age.py --dataset imagenet --dataroot /path/to/imagenet_dir/ --save_dir --image_size 32 --save_dir ${pdir} --lr 0.0002 --nz 128 --netG dcgan32px --netE dcgan32px --nepoch 6 --drop_lr 3 --e_updates '1;KL_fake:1,KL_real:1,match_z:0,match_x:10' --g_updates '2;KL_fake:1,match_z:2000,match_x:0' --workers 12
```

It can be beneficial to switch to `256` batch size after several epochs.

### Cifar10

```
python age.py --dataset cifar10 --image_size 32 --save_dir --lr 0.0002 --nz 128 --netG dcgan32px --netE dcgan32px --nepoch 150 --drop_lr 40 --e_updates '1;KL_fake:1,KL_real:1,match_z:0,match_x:10' --g_updates '2;KL_fake:1,match_z:1000,match_x:0'
```

---------------------

Tested with python 2.7.

Implementation is based on pyTorch [DCGAN code](https://github.com/pytorch/examples/tree/master/dcgan).

# Citation

If you found this code useful please cite our paper

```
@inproceedings{DBLP:conf/aaai/UlyanovVL18,
author = {Dmitry Ulyanov and
Andrea Vedaldi and
Victor S. Lempitsky},
title = {It Takes (Only) Two: Adversarial Generator-Encoder Networks},
booktitle = {{AAAI}},
publisher = {{AAAI} Press},
year = {2018}
}
```