Building blocks for uncertainty-aware cycle consistency presented at NeurIPS'21.

Overview

UncertaintyAwareCycleConsistency

This repository provides the building blocks and the API for the work presented in the NeurIPS'21 paper Robustness via Uncertainty-aware Cycle Consistency. Translation methods often learn deterministic mappings without explicitly modelling the robustness to outliers or predictive uncertainty, leading to performance degradation when encountering unseen perturbations at test time. To address this, we propose a method based on Uncertainty-aware Generalized Adaptive Cycle Consistency (UGAC), which models the per-pixel residual by generalized Gaussian distribution, capable of modelling heavy-tailed distributions.

Requirements

python >= 3.6.10
pytorch >= 1.6.0
jupyter lab
torchio
scikit-image
scikit-learn

The structure of the repository is as follows:

root
 |-ckpt/ (will save all the checkpoints)
 |-data/ (save your data and related script)
 |-src/ (contains all the source code)
    |-ds.py 
    |-networks.py
    |-utils.py
    |-losses.py

Preparing Datasets

To prepare your datasets to use with this repo, place the root directory of the dataset in data/. The recommended way to structure your data is shown below.

data/
    |-Dataset_1/
        |-A/
            |-image1.png
            |-image2.png
            |-image3.png
            |-...
        |-B/
            |-image1.png
            |-image2.png
            |-image3.png
            |-...

Note the images need not be paired. The python script src/ds.py provides the PyTorch Dataset class to read such a dataset, used as explained below.

class Images_w_nameList(data.Dataset):
    '''
    can act as supervised or un-supervised based on flists
    '''
    def __init__(self, root1, root2, flist1, flist2, transform=None):

Here root1 and root2 represents the root directory for domain A and B, respectively. flist1 and flist2 contain image names for domain A and domain B. Note, if flist1 and flist2 are aligned then dataset will load paired images. To use it as unsupervised dataset loader ensure that flist1 and flist2 are not aligned.

Learning models with uncertainty

src/networks.py provides the generator and discriminator architectures.

src/utils.py provides the training API train_UGAC. The API is to train a pair of GANs, with the generators modified to predict the parameters of the generalized Gaussian distribution GGD ($\alpha$, $\beta$, $\mu$), as depicted in the above figure.

An example command to use the first API is:

#first instantiate the generators and discriminators
netG_A = CasUNet_3head(3,3)
netD_A = NLayerDiscriminator(3, n_layers=4)
netG_B = CasUNet_3head(3,3)
netD_B = NLayerDiscriminator(3, n_layers=4)

netG_A, netD_A, netG_B, netD_B = train_UGAC(
    netG_A, netG_B,
    netD_A, netD_B,
    train_loader,
    dtype=torch.cuda.FloatTensor,
    device='cuda',
    num_epochs=10,
    init_lr=1e-5,
    ckpt_path='../ckpt/ugac',
    list_of_hp = [1, 0.015, 0.01, 0.001, 1, 0.015, 0.01, 0.001, 0.05, 0.05, 0.01],
)

This will save checkpoints in ckpt/ named as ugac_eph*.pth. The arguement list_of_hp is a list of all the hyper-parameters representing weights of different weigths in the loss function.

Apart from the code in this repository, we also use the code from many other repositories like this, this, and this.

Bibtex

If you find the bits from this project helpful, please cite the following works:

Owner
EML Tübingen
Explainable Machine Learning group at University of Tübingen
EML Tübingen
Implementation of ICCV 2021 oral paper -- A Novel Self-Supervised Learning for Gaussian Mixture Model

SS-GMM Implementation of ICCV 2021 oral paper -- Self-Supervised Image Prior Learning with GMM from a Single Noisy Image with supplementary material R

HUST-The Tan Lab 4 Dec 05, 2022
Official implementation of FCL-taco2: Fast, Controllable and Lightweight version of Tacotron2 @ ICASSP 2021

FCL-Taco2: Towards Fast, Controllable and Lightweight Text-to-Speech synthesis (ICASSP 2021) Paper | Demo Block diagram of FCL-taco2, where the decode

Disong Wang 39 Sep 28, 2022
Official code repository for the EMNLP 2021 paper

Integrating Visuospatial, Linguistic and Commonsense Structure into Story Visualization PyTorch code for the EMNLP 2021 paper "Integrating Visuospatia

Adyasha Maharana 23 Dec 19, 2022
Code Repo for the ACL21 paper "Common Sense Beyond English: Evaluating and Improving Multilingual LMs for Commonsense Reasoning"

Common Sense Beyond English: Evaluating and Improving Multilingual LMs for Commonsense Reasoning This is the Github repository of our paper, "Common S

INK Lab @ USC 19 Nov 30, 2022
This repository contains code accompanying the paper "An End-to-End Chinese Text Normalization Model based on Rule-Guided Flat-Lattice Transformer"

FlatTN This repository contains code accompanying the paper "An End-to-End Chinese Text Normalization Model based on Rule-Guided Flat-Lattice Transfor

THUHCSI 74 Nov 28, 2022
Meta Learning Backpropagation And Improving It (VSML)

Meta Learning Backpropagation And Improving It (VSML) This is research code for the NeurIPS 2021 publication Kirsch & Schmidhuber 2021. Many concepts

Louis Kirsch 22 Dec 21, 2022
Good Semi-Supervised Learning That Requires a Bad GAN

Good Semi-Supervised Learning that Requires a Bad GAN This is the code we used in our paper Good Semi-supervised Learning that Requires a Bad GAN Ziha

Zhilin Yang 177 Dec 12, 2022
Trafffic prediction analysis using hybrid models - Machine Learning

Hybrid Machine learning Model Clone the Repository Create a new Directory as assests and download the model from the below link Model Link To Start th

1 Feb 08, 2022
Code for the paper "Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds" (ICCV 2021)

Spatio-temporal Self-Supervised Representation Learning for 3D Point Clouds This is the official code implementation for the paper "Spatio-temporal Se

Hesper 63 Jan 05, 2023
RIFE - Real-Time Intermediate Flow Estimation for Video Frame Interpolation

RIFE - Real-Time Intermediate Flow Estimation for Video Frame Interpolation YouTube | BiliBili 16X interpolation results from two input images: Introd

旷视天元 MegEngine 28 Dec 09, 2022
[CVPR2021 Oral] UP-DETR: Unsupervised Pre-training for Object Detection with Transformers

UP-DETR: Unsupervised Pre-training for Object Detection with Transformers This is the official PyTorch implementation and models for UP-DETR paper: @a

dddzg 430 Dec 23, 2022
This is the official Pytorch implementation of "Lung Segmentation from Chest X-rays using Variational Data Imputation", Raghavendra Selvan et al. 2020

README This is the official Pytorch implementation of "Lung Segmentation from Chest X-rays using Variational Data Imputation", Raghavendra Selvan et a

Raghav 42 Dec 15, 2022
DeepHyper: Scalable Asynchronous Neural Architecture and Hyperparameter Search for Deep Neural Networks

What is DeepHyper? DeepHyper is a software package that uses learning, optimization, and parallel computing to automate the design and development of

DeepHyper Team 214 Jan 08, 2023
Instant-nerf-pytorch - NeRF trained SUPER FAST in pytorch

instant-nerf-pytorch This is WORK IN PROGRESS, please feel free to contribute vi

94 Nov 22, 2022
Tidy interface to polars

tidypolars tidypolars is a data frame library built on top of the blazingly fast polars library that gives access to methods and functions familiar to

Mark Fairbanks 144 Jan 08, 2023
A Python package to process & model ChEMBL data.

insilico: A Python package to process & model ChEMBL data. ChEMBL is a manually curated chemical database of bioactive molecules with drug-like proper

Steven Newton 0 Dec 09, 2021
The code is for the paper "A Self-Distillation Embedded Supervised Affinity Attention Model for Few-Shot Segmentation"

SD-AANet The code is for the paper "A Self-Distillation Embedded Supervised Affinity Attention Model for Few-Shot Segmentation" [arxiv] Overview confi

cv516Buaa 9 Nov 07, 2022
Deep universal probabilistic programming with Python and PyTorch

Getting Started | Documentation | Community | Contributing Pyro is a flexible, scalable deep probabilistic programming library built on PyTorch. Notab

7.7k Dec 30, 2022
U-Net implementation in PyTorch for FLAIR abnormality segmentation in brain MRI

U-Net for brain segmentation U-Net implementation in PyTorch for FLAIR abnormality segmentation in brain MRI based on a deep learning segmentation alg

562 Jan 02, 2023
Python Implementation of algorithms in Graph Mining, e.g., Recommendation, Collaborative Filtering, Community Detection, Spectral Clustering, Modularity Maximization, co-authorship networks.

Graph Mining Author: Jiayi Chen Time: April 2021 Implemented Algorithms: Network: Scrabing Data, Network Construbtion and Network Measurement (e.g., P

Jiayi Chen 3 Mar 03, 2022