# Image Classification on Oxford-IIIT Pet (mean per class accuracy) Source: [Image Classification on Oxford-IIIT Pet (mean per class accuracy)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:52:13.022Z Evaluation of Image Classification performance on the Oxford-IIIT Pet dataset, specifically reporting mean per class 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 | Image Classification | | Dataset (including any stated split) | Oxford-IIIT Pet | | Metric | Mean Per Class Accuracy | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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 - [Mean Per Class Accuracy](): Higher is better; unit: SCALAR. ## Recorded results | Method | Mean Per Class Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Dom-Ad | 0.971 | Not recorded | 68fa68641e209131cea1bb57 | | Full | 0.97 | Not recorded | 68fa68641e209131cea1bb57 | | Dom-Ad | 0.968 | Not recorded | 68fa68641e209131cea1bb57 | | NNCLR + Sup | 0.949 | Not recorded | 68fa6aef1e209131cea2b2cc | | Adapters | 0.946 | Not recorded | 68fa68641e209131cea1bb57 | | Baseline | 0.945 | Not recorded | 68fa68641e209131cea1bb57 | | Sup. IN | 0.932 | Not recorded | 68fa6aef1e209131cea2b2cc | | NNCLR | 0.928 | Not recorded | 68fa6aef1e209131cea2b2cc | | NNCLR | 0.925 | Not recorded | 68fa6aef1e209131cea2b2cc | | NNCLR | 0.918 | Not recorded | 68fa6aef1e209131cea2b2cc | | Sup. IN | 0.915 | Not recorded | 68fa6aef1e209131cea2b2cc | | Sup. IN | 0.914 | Not recorded | 68fa6aef1e209131cea2b2cc | | BYOL | 0.904 | Not recorded | 68fa6aef1e209131cea2b2cc | | SimCLR | 0.846 | Not recorded | 68fa6aef1e209131cea2b2cc | ### Dom-Ad Mean Per Class Accuracy: 0.971. Result ID: 698d6c868b517eb433216711. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Inception-v3"}. This record does not assert independent reproduction. ### Full Mean Per Class Accuracy: 0.97. Result ID: 698d6c868b517eb433216710. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"JFT","Protocol":"Full"}. This record does not assert independent reproduction. ### Dom-Ad Mean Per Class Accuracy: 0.968. Result ID: 698d6c868b517eb433216712. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"AmoebaNet-B"}. This record does not assert independent reproduction. ### NNCLR + Sup Mean Per Class Accuracy: 0.949. Result ID: 698e4364ffc3901de09c2157. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Arch":"ViT-B/8"}. This record does not assert independent reproduction. ### Adapters Mean Per Class Accuracy: 0.946. Result ID: 698d6c868b517eb43321670f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"JFT","Protocol":"Adapters"}. This record does not assert independent reproduction. ### Baseline Mean Per Class Accuracy: 0.945. Result ID: 698d6c868b517eb43321670e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction. ### Sup. IN Mean Per Class Accuracy: 0.932. Result ID: 698e4364ffc3901de09c2155. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Arch":"ViT-B/8"}. This record does not assert independent reproduction. ### NNCLR Mean Per Class Accuracy: 0.928. Result ID: 698e4364ffc3901de09c2156. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Arch":"ViT-B/8"}. This record does not assert independent reproduction. ### NNCLR Mean Per Class Accuracy: 0.925. Result ID: 698e4364ffc3901de09c2154. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Arch":"ViT-B/16"}. This record does not assert independent reproduction. ### NNCLR Mean Per Class Accuracy: 0.918. Result ID: 698e4364ffc3901de09c2152. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Arch":"R50"}. This record does not assert independent reproduction. ### Sup. IN Mean Per Class Accuracy: 0.915. Result ID: 698e4364ffc3901de09c2151. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Arch":"R50"}. This record does not assert independent reproduction. ### Sup. IN Mean Per Class Accuracy: 0.914. Result ID: 698e4364ffc3901de09c2153. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Arch":"ViT-B/16"}. This record does not assert independent reproduction. ### BYOL Mean Per Class Accuracy: 0.904. Result ID: 698e4364ffc3901de09c214f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Arch":"R50"}. This record does not assert independent reproduction. ### SimCLR Mean Per Class Accuracy: 0.846. Result ID: 698e4364ffc3901de09c2150. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Arch":"R50"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.