TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

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

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

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

Models/MixMatch

MixMatch

Reported on 28 benchmarks across 2 tasks · 2 papers · 21 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision30 results

  • Image ClassificationonSTL-10
    Percentage correct· 2019-05-06
    94.41
    best: 99.64 (µ2Net+ (ViT-L/16))
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-10, 2000 Labels
    Accuracy· 2019-05-06
    92.97
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSTL-10, 1000 Labels
    Accuracy· 2019-05-06
    89.82
    best: 94.53 (NP-Match)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN, 500 Labels
    Accuracy· 2019-05-06
    96.36
    best: 96.39 (Triple-GAN-V2 (CNN-13))
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN, 2000 Labels
    Accuracy· 2019-05-06
    96.96
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-10, 1000 Labels
    Accuracy· 2019-05-06
    92.25
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-10, 500 Labels
    Accuracy· 2019-05-06
    91.35
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN, 4000 Labels
    Accuracy· 2019-05-06
    97.11
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSTL-10, 5000 Labels
    Accuracy· 2019-05-06
    94.41
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN, 250 Labels
    Accuracy· 2019-05-06
    96.22
    best: 98.04 (ShrinkMatch)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-10, 250 Labels
    Percentage error· 2019-05-06
    11.08
    best: 3.47 (SemiOccam)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonCIFAR-10, 2000 Labels
    Accuracy· 2019-05-06
    92.97
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSTL-10, 1000 Labels
    Accuracy· 2019-05-06
    89.82
    best: 94.53 (NP-Match)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSVHN, 500 Labels
    Accuracy· 2019-05-06
    96.36
    best: 96.39 (Triple-GAN-V2 (CNN-13))
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSVHN, 2000 Labels
    Accuracy· 2019-05-06
    96.96
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonCIFAR-10, 1000 Labels
    Accuracy· 2019-05-06
    92.25
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonCIFAR-10, 500 Labels
    Accuracy· 2019-05-06
    91.35
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSVHN, 4000 Labels
    Accuracy· 2019-05-06
    97.11
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSTL-10, 5000 Labels
    Accuracy· 2019-05-06
    94.41
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSVHN, 250 Labels
    Accuracy· 2019-05-06
    96.22
    best: 98.04 (ShrinkMatch)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonCIFAR-10, 250 Labels
    Percentage error· 2019-05-06
    11.08
    best: 3.47 (SemiOccam)
    SOTA
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSTL-10
    Percentage correct· 2020-01-21
    89.59
    best: 99.64 (µ2Net+ (ViT-L/16))
    FixMatch: Simplifying Semi-Supervised Learning with Consistency and ConfidencearXiv:2001.07685
  • Image ClassificationonCIFAR-10
    Percentage correct· 2019-05-06
    95.05
    best: 99.5 (ViT-H/14)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-100
    Percentage correct· uses extra data· 2019-05-06
    74.1
    best: 96.08 (EffNet-L2 (SAM))
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSTL-10
    Percentage correct· 2019-05-06
    89.82
    best: 99.64 (µ2Net+ (ViT-L/16))
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN
    Percentage error· 2019-05-06
    2.59
    best: 1 (E2E-M3)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonCIFAR-10, 4000 Labels
    Percentage error· 2019-05-06
    6.24
    best: 3.96 (SimMatch)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Image ClassificationonSVHN, 1000 labels
    Accuracy· 2019-05-06
    96.73
    best: 97.58 (EnAET)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonCIFAR-10, 4000 Labels
    Percentage error· 2019-05-06
    6.24
    best: 3.96 (SimMatch)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249
  • Semi-Supervised Image ClassificationonSVHN, 1000 labels
    Accuracy· 2019-05-06
    96.73
    best: 97.58 (EnAET)
    MixMatch: A Holistic Approach to Semi-Supervised LearningarXiv:1905.02249