# 5-way 20-shot Classification on miniImageNet (test) Source: [5-way 20-shot Classification on miniImageNet (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:21:19.336Z Evaluation of 5-way, 20-shot few-shot classification performance on the test split of the miniImageNet dataset, typically measured by accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | 5-way 20-shot classification | | Dataset (including any stated split) | miniImageNet (test) | | Metric | 5-way 20-shot Accuracy | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 12 of 12 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [5-way 20-shot Accuracy](): Higher is better; unit: SCALAR. ## Recorded results | Method | 5-way 20-shot Accuracy | Result date | Report paper | | --- | --- | --- | --- | | MAML | 0.7103 | Not recorded | 68fa684f1e209131cea1b543 | | ProtoNets | 0.7005 | Not recorded | 68fa684f1e209131cea1b543 | | ULDA-ProtoNets | 0.6431 | Not recorded | 68fa684f1e209131cea1b543 | | CACTUs-MAML | 0.6384 | Not recorded | 68fa684f1e209131cea1b543 | | ULDA-MetaOptNet | 0.6358 | Not recorded | 68fa684f1e209131cea1b543 | | CACTUs-ProtoNets | 0.6154 | Not recorded | 68fa684f1e209131cea1b543 | | UMTRA | 0.6111 | Not recorded | 68fa684f1e209131cea1b543 | | knn-nearest neighbors | 0.5644 | Not recorded | 68fa684f1e209131cea1b543 | | linear classifier | 0.5619 | Not recorded | 68fa684f1e209131cea1b543 | | MLP with dropout | 0.5271 | Not recorded | 68fa684f1e209131cea1b543 | | Training from scratch | 0.5153 | Not recorded | 68fa684f1e209131cea1b543 | | cluster matching | 0.2497 | Not recorded | 68fa684f1e209131cea1b543 | ### MAML 5-way 20-shot Accuracy: 0.7103. Result ID: 698d698d8b517eb433214d0e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Supervised (Upper Bound)","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### ProtoNets 5-way 20-shot Accuracy: 0.7005. Result ID: 698d698d8b517eb433214d0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Supervised (Upper Bound)","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### ULDA-ProtoNets 5-way 20-shot Accuracy: 0.6431. Result ID: 698d698d8b517eb433214d0b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### CACTUs-MAML 5-way 20-shot Accuracy: 0.6384. Result ID: 698d698d8b517eb433214d09. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### ULDA-MetaOptNet 5-way 20-shot Accuracy: 0.6358. Result ID: 698d698d8b517eb433214d0c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### CACTUs-ProtoNets 5-way 20-shot Accuracy: 0.6154. Result ID: 698d698d8b517eb433214d08. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### UMTRA 5-way 20-shot Accuracy: 0.6111. Result ID: 698d698d8b517eb433214d0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### knn-nearest neighbors 5-way 20-shot Accuracy: 0.5644. Result ID: 698d698d8b517eb433214d04. [Reporting paper]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### linear classifier 5-way 20-shot Accuracy: 0.5619. Result ID: 698d698d8b517eb433214d05. [Reporting paper]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### MLP with dropout 5-way 20-shot Accuracy: 0.5271. Result ID: 698d698d8b517eb433214d06. [Reporting paper]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### Training from scratch 5-way 20-shot Accuracy: 0.5153. Result ID: 698d698d8b517eb433214d03. [Reporting paper]() Recorded evaluation configuration: {"Clustering":"N/A","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. ### cluster matching 5-way 20-shot Accuracy: 0.2497. Result ID: 698d698d8b517eb433214d07. [Reporting paper]() Recorded evaluation configuration: {"Clustering":"DeepCluster","Training Protocol":"Unsupervised","Backbone":"4-layer CNN"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.