Out-of-distribution detection using the pNML regret. NeurIPS2021

Last update: Dec 02, 2022

Overview

OOD Detection

Load conda environment

conda env create -f environment.yml

or install requirements:

while read requirement; do conda install --yes $requirement || pip install $requirement; done < requirements.txt

# Download OOD data
cd bash_scripts
chmod 777 ./download_data.sh
./download_data.sh

# Download pretrained models
chmod 777 ./download_models.sh
./download_models.sh

Download imagenet30

Follow https://github.com/alinlab/CSI

Imagenet30 training set: https://drive.google.com/file/d/1B5c39Fc3haOPzlehzmpTLz6xLtGyKEy4/view

Imagenet30 testing set: https://drive.google.com/file/d/13xzVuQMEhSnBRZr-YaaO08coLU2dxAUq/view

Put and untar under ./data/Imagenet30

.
├── README.md
├── data
│   ├── Imagenet30
│   │   ├── one_class_test
│   │   ├── one_class_test.tar
│   │   ├── one_class_train
│   │   └── one_class_train.tar

Execute methods

Using the pretrained models, score ood detection

cd bash_scripts
chmod 777 ./execute_methods.sh
./execute_methods.sh

Create paper's tables

cd src
python main main_create_tables.py

Out-of-distribution detection using the pNML regret. NeurIPS2021

Related tags

Overview

OOD Detection

Download imagenet30

Execute methods

Owner

Koby Bibas

Code for the paper "A Study of Face Obfuscation in ImageNet"

Code for approximate graph reduction techniques for cardinality-based DSFM, from paper

Meta Learning Backpropagation And Improving It (VSML)

Testbed of AI Systems Quality Management

Use stochastic processes to generate samples and use them to train a fully-connected neural network based on Keras

A minimal TPU compatible Jax implementation of NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis

Generating Anime Images by Implementing Deep Convolutional Generative Adversarial Networks paper

Zsseg.baseline - Zero-Shot Semantic Segmentation

[NIPS 2021] UOTA: Improving Self-supervised Learning with Automated Unsupervised Outlier Arbitration.

Human segmentation models, training/inference code, and trained weights, implemented in PyTorch

Code for PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning

Harmonic Memory Networks for Graph Completion

Relative Human dataset, CVPR 2022

Python script for performing depth completion from sparse depth and rgb images using the msg_chn_wacv20. model in Tensorflow Lite.

Human motion synthesis using Unity3D

I-SECRET: Importance-guided fundus image enhancement via semi-supervised contrastive constraining

Code for Contrastive-Geometry Networks for Generalized 3D Pose Transfer

DCSL - Generalizable Crowd Counting via Diverse Context Style Learning

Official implementation of the Implicit Behavioral Cloning (IBC) algorithm

The PASS dataset: pretrained models and how to get the data - PASS: Pictures without humAns for Self-Supervised Pretraining