guided-diffusion

This is the codebase for Diffusion Models Beat GANS on Image Synthesis.

This repository is based on openai/improved-diffusion, with modifications for classifier conditioning and architecture improvements.

Usage

Training diffusion models is described in the parent repository. Training a classifier is similar. We assume you have put training hyperparameters into a TRAIN_FLAGS variable, and classifier hyperparameters into a CLASSIFIER_FLAGS variable. Then you can run:

mpiexec -n N python scripts/classifier_train.py --data_dir path/to/imagenet $TRAIN_FLAGS $CLASSIFIER_FLAGS

Make sure to divide the batch size in TRAIN_FLAGS by the number of MPI processes you are using.

Here are flags for training the 128x128 classifier. You can modify these for training classifiers at other resolutions:

TRAIN_FLAGS="--iterations 300000 --anneal_lr True --batch_size 256 --lr 3e-4 --save_interval 10000 --weight_decay 0.05"
CLASSIFIER_FLAGS="--image_size 128 --classifier_attention_resolutions 32,16,8 --classifier_depth 2 --classifier_width 128 --classifier_pool attention --classifier_resblock_updown True --classifier_use_scale_shift_norm True"

For sampling from a 128x128 classifier-guided model, 25 step DDIM:

MODEL_FLAGS="--attention_resolutions 32,16,8 --class_cond True --image_size 128 --learn_sigma True --num_channels 256 --num_heads 4 --num_res_blocks 2 --resblock_updown True --use_fp16 True --use_scale_shift_norm True"
CLASSIFIER_FLAGS="--image_size 128 --classifier_attention_resolutions 32,16,8 --classifier_depth 2 --classifier_width 128 --classifier_pool attention --classifier_resblock_updown True --classifier_use_scale_shift_norm True --classifier_scale 1.0 --classifier_use_fp16 True"
SAMPLE_FLAGS="--batch_size 4 --num_samples 50000 --timestep_respacing ddim25 --use_ddim True"
mpiexec -n N python scripts/classifier_sample.py \
    --model_path /path/to/model.pt \
    --classifier_path path/to/classifier.pt \
    $MODEL_FLAGS $CLASSIFIER_FLAGS $SAMPLE_FLAGS

To sample for 250 timesteps without DDIM, replace --timestep_respacing ddim25 to --timestep_respacing 250, and replace --use_ddim True with --use_ddim False.

This is the codebase for Diffusion Models Beat GANS on Image Synthesis.

Related tags

Overview

guided-diffusion

Usage

Owner

OpenAI

Evaluation toolkit of the informative tracking benchmark comprising 9 scenarios, 180 diverse videos, and new challenges.

PyTorch implementations of algorithms for density estimation

Code for "The Box Size Confidence Bias Harms Your Object Detector"

BYOL for Audio: Self-Supervised Learning for General-Purpose Audio Representation

Source code for the ACL-IJCNLP 2021 paper entitled "T-DNA: Taming Pre-trained Language Models with N-gram Representations for Low-Resource Domain Adaptation" by Shizhe Diao et al.

Coarse implement of the paper "A Simultaneous Denoising and Dereverberation Framework with Target Decoupling", On DNS-2020 dataset, the DNSMOS of first stage is 3.42 and second stage is 3.47.

CondNet: Conditional Classifier for Scene Segmentation

Official code repository for the publication "Latent Equilibrium: A unified learning theory for arbitrarily fast computation with arbitrarily slow neurons"

lightweight python wrapper for vowpal wabbit

Attention-guided gan for synthesizing IR images

Unsupervised Domain Adaptation for Nighttime Aerial Tracking (CVPR2022)

Code accompanying our NeurIPS 2021 traffic4cast challenge

A very impractical 3D rendering engine that runs in the python terminal.

This project aim to create multi-label classification annotation tool to boost annotation speed and make it more easier.

Zsseg.baseline - Zero-Shot Semantic Segmentation

Stitch it in Time: GAN-Based Facial Editing of Real Videos

Learning to trade under the reinforcement learning framework

A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX

Detector for Log4Shell exploitation attempts