Progressive Growing of GANs for Improved Quality, Stability, and Variation

Overview

Progressive Growing of GANs for Improved Quality, Stability, and Variation
— Official TensorFlow implementation of the ICLR 2018 paper

Tero Karras (NVIDIA), Timo Aila (NVIDIA), Samuli Laine (NVIDIA), Jaakko Lehtinen (NVIDIA and Aalto University)

Representative image
Picture: Two imaginary celebrities that were dreamed up by a random number generator.

Abstract:
We describe a new training methodology for generative adversarial networks. The key idea is to grow both the generator and discriminator progressively: starting from a low resolution, we add new layers that model increasingly fine details as training progresses. This both speeds the training up and greatly stabilizes it, allowing us to produce images of unprecedented quality, e.g., CelebA images at 10242. We also propose a simple way to increase the variation in generated images, and achieve a record inception score of 8.80 in unsupervised CIFAR10. Additionally, we describe several implementation details that are important for discouraging unhealthy competition between the generator and discriminator. Finally, we suggest a new metric for evaluating GAN results, both in terms of image quality and variation. As an additional contribution, we construct a higher-quality version of the CelebA dataset.

★★★ NEW: StyleGAN2-ADA-PyTorch is now available; see the full list of versions here ★★★

Resources

All the material, including source code, is made freely available for non-commercial use under the Creative Commons CC BY-NC 4.0 license. Feel free to use any of the material in your own work, as long as you give us appropriate credit by mentioning the title and author list of our paper.

Versions

There are two different versions of the source code. The TensorFlow version is newer and more polished, and we generally recommend it as a starting point if you are looking to experiment with our technique, build upon it, or apply it to novel datasets. The original Theano version, on the other hand, is what we used to produce all the results shown in our paper. We recommend using it if — and only if — you are looking to reproduce our exact results for benchmark datasets like CIFAR-10, MNIST-RGB, and CelebA.

The main differences are summarized in the following table:

Feature TensorFlow version Original Theano version
Branch master (this branch) original-theano-version
Multi-GPU support Yes No
FP16 mixed-precision support Yes No
Performance High Low
Training time for CelebA-HQ 2 days (8 GPUs)
2 weeks (1 GPU)
1–2 months
Repro CelebA-HQ results Yes – very close Yes – identical
Repro LSUN results Yes – very close Yes – identical
Repro CIFAR-10 results No Yes – identical
Repro MNIST mode recovery No Yes – identical
Repro ablation study (Table 1) No Yes – identical
Dataset format TFRecords HDF5
Backwards compatibility Can import networks
trained with Theano
N/A
Code quality Reasonable Somewhat messy

System requirements

  • Both Linux and Windows are supported, but we strongly recommend Linux for performance and compatibility reasons.
  • 64-bit Python 3.6 installation with numpy 1.13.3 or newer. We recommend Anaconda3.
  • One or more high-end NVIDIA Pascal or Volta GPUs with 16GB of DRAM. We recommend NVIDIA DGX-1 with 8 Tesla V100 GPUs.
  • NVIDIA driver 391.25 or newer, CUDA toolkit 9.0 or newer, cuDNN 7.1.2 or newer.
  • Additional Python packages listed in requirements-pip.txt

Importing and using pre-trained networks

All pre-trained networks found on Google Drive, as well as ones produced by the training script, are stored as Python PKL files. They can be imported using the standard pickle mechanism as long as two conditions are met: (1) The directory containing the Progressive GAN code repository must be included in the PYTHONPATH environment variable, and (2) a tf.Session() object must have been created beforehand and set as default. Each PKL file contains 3 instances of tfutil.Network:

# Import official CelebA-HQ networks.
with open('karras2018iclr-celebahq-1024x1024.pkl', 'rb') as file:
    G, D, Gs = pickle.load(file)
    # G = Instantaneous snapshot of the generator, mainly useful for resuming a previous training run.
    # D = Instantaneous snapshot of the discriminator, mainly useful for resuming a previous training run.
    # Gs = Long-term average of the generator, yielding higher-quality results than the instantaneous snapshot.

It is also possible to import networks that were produced using the Theano implementation, as long as they do not employ any features that are not natively supported by the TensorFlow version (minibatch discrimination, batch normalization, etc.). To enable Theano network import, however, you must use misc.load_pkl() in place of pickle.load():

# Import Theano versions of the official CelebA-HQ networks.
import misc
G, D, Gs = misc.load_pkl('200-celebahq-1024x1024/network-final.pkl')

Once you have imported the networks, you can call Gs.run() to produce a set of images for given latent vectors, or Gs.get_output_for() to include the generator network in a larger TensorFlow expression. For further details, please consult the example script found on Google Drive. Instructions:

  1. Pull the Progressive GAN code repository and add it to your PYTHONPATH environment variable.
  2. Install the required Python packages with pip install -r requirements-pip.txt
  3. Download import_example.py from networks/tensorflow-version/example_import_script
  4. Download karras2018iclr-celebahq-1024x1024.pkl from networks/tensorflow-version and place it in the same directory as the script.
  5. Run the script with python import_example.py
  6. If everything goes well, the script should generate 10 PNG images (img0.pngimg9.png) that match the ones found in networks/tensorflow-version/example_import_script exactly.

Preparing datasets for training

The Progressive GAN code repository contains a command-line tool for recreating bit-exact replicas of the datasets that we used in the paper. The tool also provides various utilities for operating on the datasets:

usage: dataset_tool.py [-h] <command> ...

    display             Display images in dataset.
    extract             Extract images from dataset.
    compare             Compare two datasets.
    create_mnist        Create dataset for MNIST.
    create_mnistrgb     Create dataset for MNIST-RGB.
    create_cifar10      Create dataset for CIFAR-10.
    create_cifar100     Create dataset for CIFAR-100.
    create_svhn         Create dataset for SVHN.
    create_lsun         Create dataset for single LSUN category.
    create_celeba       Create dataset for CelebA.
    create_celebahq     Create dataset for CelebA-HQ.
    create_from_images  Create dataset from a directory full of images.
    create_from_hdf5    Create dataset from legacy HDF5 archive.

Type "dataset_tool.py <command> -h" for more information.

The datasets are represented by directories containing the same image data in several resolutions to enable efficient streaming. There is a separate *.tfrecords file for each resolution, and if the dataset contains labels, they are stored in a separate file as well:

> python dataset_tool.py create_cifar10 datasets/cifar10 ~/downloads/cifar10
> ls -la datasets/cifar10
drwxr-xr-x  2 user user         7 Feb 21 10:07 .
drwxrwxr-x 10 user user        62 Apr  3 15:10 ..
-rw-r--r--  1 user user   4900000 Feb 19 13:17 cifar10-r02.tfrecords
-rw-r--r--  1 user user  12350000 Feb 19 13:17 cifar10-r03.tfrecords
-rw-r--r--  1 user user  41150000 Feb 19 13:17 cifar10-r04.tfrecords
-rw-r--r--  1 user user 156350000 Feb 19 13:17 cifar10-r05.tfrecords
-rw-r--r--  1 user user   2000080 Feb 19 13:17 cifar10-rxx.labels

The create_* commands take the standard version of a given dataset as input and produce the corresponding *.tfrecords files as output. Additionally, the create_celebahq command requires a set of data files representing deltas with respect to the original CelebA dataset. These deltas (27.6GB) can be downloaded from datasets/celeba-hq-deltas.

Note about module versions: Some of the dataset commands require specific versions of Python modules and system libraries (e.g. pillow, libjpeg), and they will give an error if the versions do not match. Please heed the error messages — there is no way to get the commands to work other than installing these specific versions.

Training networks

Once the necessary datasets are set up, you can proceed to train your own networks. The general procedure is as follows:

  1. Edit config.py to specify the dataset and training configuration by uncommenting/editing specific lines.
  2. Run the training script with python train.py.
  3. The results are written into a newly created subdirectory under config.result_dir
  4. Wait several days (or weeks) for the training to converge, and analyze the results.

By default, config.py is configured to train a 1024x1024 network for CelebA-HQ using a single-GPU. This is expected to take about two weeks even on the highest-end NVIDIA GPUs. The key to enabling faster training is to employ multiple GPUs and/or go for a lower-resolution dataset. To this end, config.py contains several examples for commonly used datasets, as well as a set of "configuration presets" for multi-GPU training. All of the presets are expected to yield roughly the same image quality for CelebA-HQ, but their total training time can vary considerably:

  • preset-v1-1gpu: Original config that was used to produce the CelebA-HQ and LSUN results shown in the paper. Expected to take about 1 month on NVIDIA Tesla V100.
  • preset-v2-1gpu: Optimized config that converges considerably faster than the original one. Expected to take about 2 weeks on 1xV100.
  • preset-v2-2gpus: Optimized config for 2 GPUs. Takes about 1 week on 2xV100.
  • preset-v2-4gpus: Optimized config for 4 GPUs. Takes about 3 days on 4xV100.
  • preset-v2-8gpus: Optimized config for 8 GPUs. Takes about 2 days on 8xV100.

For reference, the expected output of each configuration preset for CelebA-HQ can be found in networks/tensorflow-version/example_training_runs

Other noteworthy config options:

  • fp16: Enable FP16 mixed-precision training to reduce the training times even further. The actual speedup is heavily dependent on GPU architecture and cuDNN version, and it can be expected to increase considerably in the future.
  • BENCHMARK: Quickly iterate through the resolutions to measure the raw training performance.
  • BENCHMARK0: Same as BENCHMARK, but only use the highest resolution.
  • syn1024rgb: Synthetic 1024x1024 dataset consisting of just black images. Useful for benchmarking.
  • VERBOSE: Save image and network snapshots very frequently to facilitate debugging.
  • GRAPH and HIST: Include additional data in the TensorBoard report.

Analyzing results

Training results can be analyzed in several ways:

  • Manual inspection: The training script saves a snapshot of randomly generated images at regular intervals in fakes*.png and reports the overall progress in log.txt.
  • TensorBoard: The training script also exports various running statistics in a *.tfevents file that can be visualized in TensorBoard with tensorboard --logdir <result_subdir>.
  • Generating images and videos: At the end of config.py, there are several pre-defined configs to launch utility scripts (generate_*). For example:
    • Suppose you have an ongoing training run titled 010-pgan-celebahq-preset-v1-1gpu-fp32, and you want to generate a video of random interpolations for the latest snapshot.
    • Uncomment the generate_interpolation_video line in config.py, replace run_id=10, and run python train.py
    • The script will automatically locate the latest network snapshot and create a new result directory containing a single MP4 file.
  • Quality metrics: Similar to the previous example, config.py also contains pre-defined configs to compute various quality metrics (Sliced Wasserstein distance, Fréchet inception distance, etc.) for an existing training run. The metrics are computed for each network snapshot in succession and stored in metric-*.txt in the original result directory.
Official implementation of VaxNeRF (Voxel-Accelearated NeRF).

VaxNeRF Paper | Google Colab This is the official implementation of VaxNeRF (Voxel-Accelearated NeRF). VaxNeRF provides very fast training and slightl

naruya 132 Nov 21, 2022
CVPR '21: In the light of feature distributions: Moment matching for Neural Style Transfer

In the light of feature distributions: Moment matching for Neural Style Transfer (CVPR 2021) This repository provides code to recreate results present

Nikolai Kalischek 49 Oct 13, 2022
COPA-SSE contains crowdsourced explanations for the Balanced COPA dataset

COPA-SSE Repository for COPA-SSE: Semi-Structured Explanations for Commonsense Reasoning. COPA-SSE contains crowdsourced explanations for the Balanced

Ana Brassard 5 Jul 31, 2022
An example of Scatterbrain implementation (combining local attention and Performer)

An example of Scatterbrain implementation (combining local attention and Performer)

HazyResearch 97 Jan 02, 2023
Code release for NeurIPS 2020 paper "Co-Tuning for Transfer Learning"

CoTuning Official implementation for NeurIPS 2020 paper Co-Tuning for Transfer Learning. [News] 2021/01/13 The COCO 70 dataset used in the paper is av

THUML @ Tsinghua University 35 Sep 23, 2022
Contains a bunch of different python programm tasks

py_tasks Contains a bunch of different python programm tasks Armstrong.py - calculate Armsrong numbers in range from 0 to n with / without cache and c

Dmitry Chmerenko 1 Dec 17, 2021
The repository for our EMNLP 2021 paper "Finnish Dialect Identification: The Effect of Audio and Text"

Finnish Dialect Identification The repository for our EMNLP 2021 paper "Finnish Dialect Identification: The Effect of Audio and Text". We present a te

Rootroo Ltd 2 Dec 25, 2021
Fastquant - Backtest and optimize your trading strategies with only 3 lines of code!

fastquant 🤓 Bringing backtesting to the mainstream fastquant allows you to easily backtest investment strategies with as few as 3 lines of python cod

Lorenzo Ampil 1k Dec 29, 2022
Code for T-Few from "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning"

T-Few This repository contains the official code for the paper: "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learni

220 Dec 31, 2022
A Simple LSTM-Based Solution for "Heartbeat Signal Classification and Prediction" in Tianchi

LSTM-Time-Series-Prediction A Simple LSTM-Based Solution for "Heartbeat Signal Classification and Prediction" in Tianchi Contest. The Link of the Cont

KevinCHEN 1 Jun 13, 2022
Conditional Gradients For The Approximately Vanishing Ideal

Conditional Gradients For The Approximately Vanishing Ideal Code for the paper: Wirth, E., and Pokutta, S. (2022). Conditional Gradients for the Appro

IOL Lab @ ZIB 0 May 25, 2022
Matlab Python Heuristic Battery Opt - SMOP conversion and manual conversion

SMOP is Small Matlab and Octave to Python compiler. SMOP translates matlab to py

Tom Xu 1 Jan 12, 2022
A Unified Framework and Analysis for Structured Knowledge Grounding

UnifiedSKG 📚 : Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models Code for paper UnifiedSKG: Unifying and Mu

HKU NLP Group 370 Dec 21, 2022
This is the code repository for the paper "Identification of the Generalized Condorcet Winner in Multi-dueling Bandits" (NeurIPS 2021).

Code Repository for the Paper "Identification of the Generalized Condorcet Winner in Multi-dueling Bandits" (To appear in: Proceedings of NeurIPS20

1 Oct 03, 2022
Project page for our ICCV 2021 paper "The Way to my Heart is through Contrastive Learning"

The Way to my Heart is through Contrastive Learning: Remote Photoplethysmography from Unlabelled Video This is the official project page of our ICCV 2

36 Jan 06, 2023
Code for Contrastive-Geometry Networks for Generalized 3D Pose Transfer

CGTransformer Code for our AAAI 2022 paper "Contrastive-Geometry Transformer network for Generalized 3D Pose Transfer" Contrastive-Geometry Transforme

18 Jun 28, 2022
[NeurIPS 2021] A weak-shot object detection approach by transferring semantic similarity and mask prior.

TransMaS This repository is the official pytorch implementation of the following paper: NIPS2021 Mixed Supervised Object Detection by TransferringMask

BCMI 49 Jul 27, 2022
RuDOLPH: One Hyper-Modal Transformer can be creative as DALL-E and smart as CLIP

[Paper] [Хабр] [Model Card] [Colab] [Kaggle] RuDOLPH 🦌 🎄 ☃️ One Hyper-Modal Tr

Sber AI 230 Dec 31, 2022
SymPy-powered, Wolfram|Alpha-like answer engine totally in your browser, without backend computation

SymPy Beta SymPy Beta is a fork of SymPy Gamma. The purpose of this project is to run a SymPy-powered, Wolfram|Alpha-like answer engine totally in you

Liumeo 25 Dec 21, 2022
Finite-temperature variational Monte Carlo calculation of uniform electron gas using neural canonical transformation.

CoulombGas This code implements the neural canonical transformation approach to the thermodynamic properties of uniform electron gas. Building on JAX,

FermiFlow 9 Mar 03, 2022