Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Datasets/CIFAR-100

CIFAR-100

ImagesIntroduced 2009-04-08

The CIFAR-100 dataset (Canadian Institute for Advanced Research, 100 classes) is a subset of the Tiny Images dataset and consists of 60000 32x32 color images. The 100 classes in the CIFAR-100 are grouped into 20 superclasses. There are 600 images per class. Each image comes with a "fine" label (the class to which it belongs) and a "coarse" label (the superclass to which it belongs). There are 500 training images and 100 testing images per class.

The criteria for deciding whether an image belongs to a class were as follows:

The class name should be high on the list of likely answers to the question “What is in this picture?”
The image should be photo-realistic. Labelers were instructed to reject line drawings.
The image should contain only one prominent instance of the object to which the class refers.
The object may be partially occluded or seen from an unusual viewpoint as long as its identity is still clear to the labeler.

Source: https://www.cs.toronto.edu/~kriz/cifar.html Image Source: https://www.cs.toronto.edu/~kriz/cifar.html

Benchmarks

Adversarial Attack/Attack: AutoAttack Adversarial Defense/autoattack Adversarial Defense/Accuracy Adversarial Robustness/Clean Accuracy Adversarial Robustness/AutoAttacked Accuracy AutoML/Percentage Error AutoML/FLOPS AutoML/PARAMS AutoML/Search Time (GPU days)AutoML/Accuracy (% )Class Incremental Learning/Last Accuracy Class Incremental Learning/Average Accuracy Classification/Accuracy Classification/Expected Calibration Error Conditional Image Generation/FID Conditional Image Generation/Inception Score Conditional Image Generation/Intra-FID Continual Learning/Last Accuracy Continual Learning/Average Accuracy Document Text Classification/Test Accuracy Federated Learning/ACC@1-500 Federated Learning/ACC@1-100Clients Federated Learning/ACC@1-50Clients Federated Learning/ACC@1-10Clients Federated Learning/ACC@5-100Clients Image Classification/Percentage correct Image Classification/PARAMS Image Classification/Accuracy Image Classification/Top 1 Accuracy Image Classification/Test Accuracy Image Classification/All accuracy (10% Labeled)Image Classification/Novel accuracy (10% Labeled)Image Classification/Seen accuracy (10% Labeled)Image Classification/All accuracy (50% Labeled)Image Classification/Novel accuracy (50% Labeled)Image Classification/Seen accuracy (50% Labeled)Image Clustering/Accuracy Image Clustering/NMI Image Clustering/ARI Image Clustering/Train Set Image Clustering/Backbone Image Generation/FID Image Generation/Inception Score Image Generation/Model Size (MB)Image Generation/Intra-FID Knowledge Distillation/Top-1 Accuracy (%)Network Pruning/Accuracy Network Pruning/GFLOPs Network Pruning/Inference Time (ms)Neural Architecture Search/Percentage Error Neural Architecture Search/FLOPS Neural Architecture Search/PARAMS Neural Architecture Search/Search Time (GPU days)Neural Architecture Search/Accuracy (% )Out-of-Distribution Detection/FPR95 Out-of-Distribution Detection/AUROC Quantization/CIFAR-100 W4A4 Top-1 Accuracy Quantization/CIFAR-100 W5A5 Top-1 Accuracy Quantization/CIFAR-100 W6A6 Top-1 Accuracy Quantization/CIFAR-100 W8A8 Top-1 Accuracy Self-Supervised Learning/Top-1 Accuracy Semi-Supervised Image Classification/All accuracy (10% Labeled)Semi-Supervised Image Classification/Novel accuracy (10% Labeled)Semi-Supervised Image Classification/Seen accuracy (10% Labeled)Semi-Supervised Image Classification/All accuracy (50% Labeled)Semi-Supervised Image Classification/Novel accuracy (50% Labeled)Semi-Supervised Image Classification/Seen accuracy (50% Labeled)Stochastic Optimization/Accuracy (max)Stochastic Optimization/Accuracy (mean)Zero-Shot Learning/Accuracy Zero-Shot Learning/Accuarcy

Related Benchmarks

CIFAR-100 (10000 Labels, ImageNet-100 Unlabeled)/Image Classification/Accuracy CIFAR-100 (10000 Labels, ImageNet-100 Unlabeled)/Semi-Supervised Image Classification/Accuracy CIFAR-100 (250 Labels, ImageNet-100 Unlabeled)/Image Classification/Accuarcy CIFAR-100 (250 Labels, ImageNet-100 Unlabeled)/Semi-Supervised Image Classification/Accuarcy CIFAR-100 (400 Labels, ImageNet-100 Unlabeled)/Image Classification/Accuracy CIFAR-100 (400 Labels, ImageNet-100 Unlabeled)/Semi-Supervised Image Classification/Accuracy CIFAR-100 (alpha=0, 10 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=0, 20 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=0, 20 clients per round)/Image Classification/ACC@1-100Clients CIFAR-100 (alpha=0, 5 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=0.5, 10 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=0.5, 20 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=0.5, 5 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=1000, 10 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=1000, 20 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (alpha=1000, 5 clients per round)/Federated Learning/ACC@1-100Clients CIFAR-100 (partial ratio 0.01)/Partial Label Learning/Accuracy CIFAR-100 (partial ratio 0.05)/Partial Label Learning/Accuracy CIFAR-100 (partial ratio 0.1)/Partial Label Learning/Accuracy CIFAR-100 - 40 classes + 60 steps of 1 class (Exemplar-free)/Incremental Learning/Average Incremental Accuracy CIFAR-100 - 50 classes + 10 steps of 5 classes/Class Incremental Learning/Final Accuracy CIFAR-100 - 50 classes + 10 steps of 5 classes/Continual Learning/Final Accuracy CIFAR-100 - 50 classes + 10 steps of 5 classes/Incremental Learning/Average Incremental Accuracy CIFAR-100 - 50 classes + 2 steps of 25 classes/Incremental Learning/Average Incremental Accuracy CIFAR-100 - 50 classes + 25 steps of 2 classes/Incremental Learning/Average Incremental Accuracy CIFAR-100 - 50 classes + 5 steps of 10 classes/Class Incremental Learning/Final Accuracy CIFAR-100 - 50 classes + 5 steps of 10 classes/Continual Learning/Final Accuracy CIFAR-100 - 50 classes + 5 steps of 10 classes/Incremental Learning/Average Incremental Accuracy CIFAR-100 - 50 classes + 5 steps of 10 classes/Incremental Learning/Final Accuracy CIFAR-100 - 50 classes + 50 steps of 1 class/Incremental Learning/Average Incremental Accuracy CIFAR-100 AlexNet - 300 Epoch/Continual Learning/Accuracy CIFAR-100 ResNet-18 - 300 Epochs/Continual Learning/Accuracy CIFAR-100 WRN-28-10 - 200 Epochs/Stochastic Optimization/Accuracy CIFAR-100 vs CIFAR-10/Out-of-Distribution Detection/AUPR CIFAR-100 vs CIFAR-10/Out-of-Distribution Detection/AUROC CIFAR-100 vs Gaussian/Out-of-Distribution Detection/AUROC CIFAR-100 vs ImageNet (C)/Out-of-Distribution Detection/AUROC CIFAR-100 vs ImageNet (R)/Out-of-Distribution Detection/AUROC CIFAR-100 vs LSUN (C)/Out-of-Distribution Detection/AUROC CIFAR-100 vs LSUN (R)/Out-of-Distribution Detection/AUROC CIFAR-100 vs SVHN/Out-of-Distribution Detection/AUROC CIFAR-100 vs Uniform/Out-of-Distribution Detection/AUROC CIFAR-100 vs iSUN/Out-of-Distribution Detection/AUROC CIFAR-100, 1000 Labels/Image Classification/Accuracy CIFAR-100, 1000 Labels/Image Classification/Percentage correct CIFAR-100, 1000 Labels/Semi-Supervised Image Classification/Percentage correct CIFAR-100, 200 Labels/Image Classification/Percentage error CIFAR-100, 200 Labels/Semi-Supervised Image Classification/Percentage error CIFAR-100, 2500 Labels/Image Classification/Percentage error CIFAR-100, 2500 Labels/Semi-Supervised Image Classification/Percentage error CIFAR-100, 40% Symmetric Noise/Image Classification/Percentage correct CIFAR-100, 400 Labels/Image Classification/Percentage error CIFAR-100, 400 Labels/Semi-Supervised Image Classification/Percentage error CIFAR-100, 4000 Labels/Image Classification/Accuracy CIFAR-100, 4000 Labels/Semi-Supervised Image Classification/Accuracy CIFAR-100, 5000 Labels/Image Classification/Accuracy (%)CIFAR-100, 5000 Labels/Semi-Supervised Image Classification/Accuracy (%)CIFAR-100, 5000Labels/Image Classification/Percentage correct CIFAR-100, 5000Labels/Semi-Supervised Image Classification/Percentage correct CIFAR-100, 60% Symmetric Noise/Image Classification/Percentage correct CIFAR-100-B0(5steps of 20 classes)/Incremental Learning/Average Incremental Accuracy CIFAR-100-LT (ρ=10)/Few-Shot Image Classification/Error Rate CIFAR-100-LT (ρ=10)/Generalized Few-Shot Classification/Error Rate CIFAR-100-LT (ρ=10)/Generalized Few-Shot Learning/Error Rate CIFAR-100-LT (ρ=10)/Image Classification/Error Rate CIFAR-100-LT (ρ=10)/Long-tail Learning/Error Rate CIFAR-100-LT (ρ=100)/Few-Shot Image Classification/Error Rate CIFAR-100-LT (ρ=100)/Generalized Few-Shot Classification/Error Rate CIFAR-100-LT (ρ=100)/Generalized Few-Shot Learning/Error Rate CIFAR-100-LT (ρ=100)/Image Classification/Error Rate CIFAR-100-LT (ρ=100)/Long-tail Learning/Error Rate CIFAR-100-LT (ρ=200)/Few-Shot Image Classification/Error Rate CIFAR-100-LT (ρ=200)/Generalized Few-Shot Classification/Error Rate CIFAR-100-LT (ρ=200)/Generalized Few-Shot Learning/Error Rate CIFAR-100-LT (ρ=200)/Image Classification/Error Rate CIFAR-100-LT (ρ=200)/Long-tail Learning/Error Rate CIFAR-100-LT (ρ=50)/Few-Shot Image Classification/Error Rate CIFAR-100-LT (ρ=50)/Generalized Few-Shot Classification/Error Rate CIFAR-100-LT (ρ=50)/Generalized Few-Shot Learning/Error Rate CIFAR-100-LT (ρ=50)/Image Classification/Error Rate CIFAR-100-LT (ρ=50)/Long-tail Learning/Error Rate CIFAR-100C/Domain Adaptation/Accuracy CIFAR-100C/Domain Generalization/Accuracy CIFAR-100C/Image Classification/Percentage correct CIFAR-100N/Document Text Classification/Accuracy (mean)CIFAR-100N/Image Classification/Accuracy (mean)cifar-100, 10000 Labels/Image Classification/Percentage error cifar-100, 10000 Labels/Semi-Supervised Image Classification/Percentage error

Statistics

Papers: 9,045
Benchmarks: 71

Links

Tasks

Adversarial Attack Adversarial Defense Adversarial Robustness Anomaly Detection AutoML Bayesian Inference Binarization Class Incremental Learning Classification Classification with Binary Neural Network Classification with Binary Weight Network Classifier calibration Conditional Image Generation Continual Learning Data Free Quantization Dataset Distillation - 1IPC Document Text Classification Federated Learning Few-Shot Class-Incremental Learning Few-Shot Image Classification Image Classification Image Classification with Label Noise Image Clustering Image Generation Incremental Learning Knowledge Distillation Learning with coarse labels Learning with noisy labels Long-tail Learning Network Pruning Neural Architecture Search Non-exemplar-based Class Incremental Learning Novel Class Discovery Open-World Semi-Supervised Learning Out-of-Distribution Detection Personalized Federated Learning Provable Adversarial Defense Quantization Self-Supervised Learning Semi-Supervised Image Classification Small Data Image Classification Sparse Learning and binarization Stochastic Optimization Transductive Zero-Shot Classification Variational Inference Zero-Shot Learning class-incremental learning