Code for paper "Adversarial score matching and improved sampling for image generation"

Overview

Adversarial score matching and improved sampling for image generation

This repo contains the official implementation for the ICLR 2021 paper Adversarial score matching and improved sampling for image generation. It it a highly extended version of the original repo on score matching.

Discussion and more samples at https://ajolicoeur.wordpress.com/adversarial-score-matching-and-consistent-sampling.


Denoising score matching with Annealed Langevin Sampling (DSM-ALS) is a recent approach to generative modeling. Despite the convincing visual quality of samples, this method appears to perform worse than Generative Adversarial Networks (GANs) under the Frechet Inception Distance, a popular metric forgenerative models. We show that this apparent gap vanishes when denoising thefinal Langevin samples using the score network. In addition, we propose two improvements to DSM-ALS: 1) Consistent Annealed Sampling as a more stable alternative to Annealed Langevin Sampling, and 2) a hybrid training formulation, composed of both denoising score matching and adversarial objectives. By combining both of these techniques and exploring different network architectures, we elevate score matching methods and obtain results competitive with state-of-the-art image generation on CIFAR-10

Adversarial LSUN-Churches


Citation

If you find this code useful please cite us in your work:

@inproceedings{
  jolicoeur-martineau2021adversarial,
  title={Adversarial score matching and improved sampling for image generation},
  author={Alexia Jolicoeur-Martineau and R{\'e}mi Pich{\'e}-Taillefer and Ioannis Mitliagkas and Remi Tachet des Combes},
  booktitle={International Conference on Learning Representations},
  year={2021},
  url={https://openreview.net/forum?id=eLfqMl3z3lq}
}

Setup

Needed

Hyperparameter choice

  • Use "main.py --compute_approximate_sigma_max" to choose model.sigma_begin based on the current dataset (based on Technique 1 from https://arxiv.org/abs/2006.09011)
  • Use calculate_number_of_steps.R to choose model.num_classes (based on Technique 2 from https://arxiv.org/abs/2006.09011)
  • tune sampling.step_lr manually for consistent or non-consistent with n_sigma=1 (see Appendix B for how to extrapolate to n_sigma > 1 from the step_lr at n_sigma = 1)
  • Everything else can be left to default

To Replicate paper

To train a non-adversarial score network

python main.py --config cifar10_9999ema.yml --doc cifar10_bs128_L2_9999ema --ni

Log files will be saved in <exp>/logs/cifar10_bs128_L2_9999ema.

To train an adversarial score network

python main.py --config cifar10_9999ema.yml --doc cifar10_bs128_L2_9999ema_adam0_9_adamD-5_9_LSGAN_ --ni  --adam --adam_beta 0 .9 --D_adam --D_adam_beta -.5 .9 --adversarial

Log files will be saved in <exp>/logs/cifar10_bs128_L2_9999ema_adam0_9_adamD-5_9_LSGAN_.

To sample from a pre-trained score network (ex: cifar10_bs128_L2_9999ema, consistent, nsigma=1)

python main.py --sample --config cifar10_9999ema.yml -i cifar10_bs128_L2_9999ema --ni --consistent --nsigma 1 --step_lr 5.6e-6 --batch_size 100 --begin_ckpt 250000

Samples will be saved in <exp>/image_samples/cifar10_bs128_L2_9999ema.

To compute the FID for a range of checkpoints from a pre-trained score network (ex: cifar10_bs128_L2_9999ema, at 100k to 300k iterations)

python main.py --fast_fid --config cifar10_9999ema.yml -i cifar10_bs128_L2_9999ema --ni --consistent --nsigma 1 --step_lr 5.6e-6 --batch_size 4000 --fid_num_samples 10000 --begin_ckpt 100000 --end_ckpt 300000

FIDs will be saved in {args.fid_folder}/log_FID.txt.

Pretrained Score Network Checkpoints

Link: https://www.dropbox.com/s/dltiobdlsb2vhyo/DSM_ScoreNetwork_Pretrained.zip?dl=0

Download and unzip it to the exp folder.

FID statistics (for FID evaluation)

Link: https://www.dropbox.com/s/nhvp2tf1unxj08g/fid_stats.zip?dl=0

Download and unzip it to the exp/datasets folder.

Owner
Alexia Jolicoeur-Martineau
I'm a researcher in statistics and machine learning. I am particularly interested in GANs, denoising score-matching, and statistical divergences/metrics.
Alexia Jolicoeur-Martineau
A file-based quote bot written in Python

Let's Write a Python Quote Bot! This repository will get you started with building a quote bot in Python. It's meant to be used along with the Learnin

1 Oct 28, 2021
Jalali version of python calendar :date:

jcalendar jcalendar is Jalali implementation of Python's calendar module Status Install pip install jcalendar Documents This module almost follows Py

Iman Kermani 7 Aug 09, 2022
arweave-nft-uploader is a Python tool to improve the experience of uploading NFTs to the Arweave storage for use with the Metaplex Candy Machine.

arweave-nft-uploader arweave-nft-uploader is a Python tool to improve the experience of uploading NFTs to the Arweave storage for use with the Metaple

0xEnrico 84 Dec 26, 2022
A Telegram Bot written in Python for mirroring files on the Internet to Google Drive

No support is going to be provided of any kind, only maintaining this for vps user on request. This is a Telegram Bot written in Python for mirroring

0 Dec 26, 2021
Awslogs - AWS CloudWatch logs for Humans™

awslogs awslogs is a simple command line tool for querying groups, streams and events from Amazon CloudWatch logs. One of the most powerful features i

Jorge Bastida 4.5k Dec 30, 2022
YouTube-Discord-Bot - Discord Bot to Search YouTube

YouTube Bot Info YouTube Bot is a discord bot where you can search for anything

Riceblades11 10 Mar 05, 2022
Bot that embeds a random hysterical meme from Reddit into your text channel as an embedded message, using an API call.

Discord_Meme_Bot 🤣 Bot that embeds a random hysterical meme from Reddit into your text channel as an embedded message, using an API call. Add the bot

2 Jan 16, 2022
Amanda-A next gen powerful telegram group manager bot for manage your groups and have fun with other cool modules.

Amanda-A next gen powerful telegram group manager bot for manage your groups and have fun with other cool modules.

Team Amanda 4 Oct 21, 2022
Example of a discord bot in Python

discordbot.py Example of a discord bot in Python Requirements Python 3.8 or higher Discord Bot Setting Up Clone this repo or download the files Rename

Debert Jamie 1 Oct 23, 2021
Código para trabalho com o dataset Wine em Python

Um perceptron multicamadas (MLP) é uma rede neural artificial feedforward que gera um conjunto de saídas a partir de um conjunto de entradas. Um MLP é

Hemili Beatriz 1 Jan 08, 2022
Cogs for Red-DiscordBot

Redbot cogs for Red-DiscordBot authored by Kreusada This is my repository for Red Discord-Bot. I built these cogs because these were the features that

Kreus Amredes 26 Nov 07, 2022
An information scroller Twitter trends, news, weather for raspberry pi and Pimoroni Unicorn Hat Mini and Scroll Phat HD.

uticker An information scroller Twitter trends, news, weather for raspberry pi and Pimoroni Unicorn Hat Mini and Scroll Phat HD. Features include: Twi

kottuora 5 Oct 31, 2022
Discord Voice Call DoS

VC DoS Simple, effective Discord DM/GC voice call Denial of Service. How to Use & FAQ 1. Download the script (obviously). 2. In CMD prompt, find the l

Roover 4 Feb 28, 2022
Signs the target email up to over 1000 different mailing lists to get spammed each day.

Email Bomber Say goodbye to that email Features Signs up to over 1k different mailing lists Written in python so the program is lightweight Easy to us

Loxdr 1 Nov 30, 2021
Telegram bot to clip youtube videos

youtube-clipper-bot Telegram bot to clip youtube videos How to deploy? Create a file called config.env BOT_TOKEN: Provide your bot token generated by

Shivam Jha 11 Dec 10, 2022
Bulk convert image types with Python

Bulk Image Converter 🔥 Helper script to convert a folder's worth of images from one filetype to another, and optionally delete originals Use Setup /

1 Nov 13, 2021
Plazmix API wrapper for Python

An optimised, easy to use Plazmix API wrapper written in Python

Someone 2 Nov 16, 2021
Instagram-follower-bot - An Instagram follower bot written in Python

Instagram Follower Bot An Instagram follower bot written in Python. The bot follows the follower of which account you want. e.g. (You want to follow @

Aytaç Kaşoğlu 1 Dec 31, 2021
Fetching tweets and integrating it with Kafka and PySpark

KafkaPySpark Zookeeper bin/zookeeper-server-start.sh config/zookeeper.properties Kafka Server bin/kafka-server-start.sh config/server.properties Kafka

Priyansh 2 Dec 29, 2021
Select random winners for a Twitter giveaway

twitter_picker Select random winners for a Twitter giveaway Once the Twitter giveaway (or airdrop) is closed, assign a number to each participant. The

Michael Rawner 1 Dec 11, 2021