# Image Classification on Fashion-MNIST standard (test) Source: [Image Classification on Fashion-MNIST standard (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:32:58.447Z Evaluation of Image Classification performance on the standard test split of the Fashion-MNIST dataset, reporting accuracy as a percentage. 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) | Fashion-MNIST standard (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | WRN-28-10 + Random Erasing | 96.3 | Not recorded | 68fa696a1e209131cea21e2c | | AVG-Softmax | 95.92 | Not recorded | 68fa696a1e209131cea21e2c | | EXT-Softmax | 95.91 | Not recorded | 68fa696a1e209131cea21e2c | | WRN-28-10 | 95.9 | Not recorded | 68fa696a1e209131cea21e2c | | E2E-3M | 95.85 | Not recorded | 68fa696a1e209131cea21e2c | | Dual path network with WRN-28-10 | 95.7 | Not recorded | 68fa696a1e209131cea21e2c | | SeResNeXt101-STD | 95.41 | Not recorded | 68fa696a1e209131cea21e2c | | DENSER | 95.3 | Not recorded | 68fa696a1e209131cea21e2c | | MobileNet | 95 | Not recorded | 68fa696a1e209131cea21e2c | | CNN with optional shortcuts | 94.7 | Not recorded | 68fa696a1e209131cea21e2c | | Google AutoML | 93.9 | Not recorded | 68fa696a1e209131cea21e2c | | BitLogic | 93.81 | Not recorded | 698abb70c97a0b3e3b5dde78 | | Capsule Network | 93.6 | Not recorded | 68fa696a1e209131cea21e2c | | VGG16 | 93.5 | Not recorded | 68fa696a1e209131cea21e2c | | LeNet | 93.4 | Not recorded | 68fa696a1e209131cea21e2c | | DWN (n=2) | 89.12 | Not recorded | 698abb70c97a0b3e3b5dde78 | | DWN (n=6) | 89.01 | Not recorded | 698abb70c97a0b3e3b5dde78 | | BitLogic | 81.07 | Not recorded | 698abb70c97a0b3e3b5dde78 | ### WRN-28-10 + Random Erasing Accuracy: 96.3. Result ID: 699aab4c5698665ec4384fef. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@zhunzhong07"}. This record does not assert independent reproduction. ### AVG-Softmax Accuracy: 95.92. Result ID: 699aab4c5698665ec4384ff9. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"N/A"}. This record does not assert independent reproduction. ### EXT-Softmax Accuracy: 95.91. Result ID: 699aab4c5698665ec4384ffa. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Submitter":"N/A"}. This record does not assert independent reproduction. ### WRN-28-10 Accuracy: 95.9. Result ID: 699aab4c5698665ec4384ff0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Submitter":"@zhunzhong07"}. This record does not assert independent reproduction. ### E2E-3M Accuracy: 95.85. Result ID: 699aab4c5698665ec4384ffb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Submitter":"N/A"}. This record does not assert independent reproduction. ### Dual path network with WRN-28-10 Accuracy: 95.7. Result ID: 699aab4c5698665ec4384ff1. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@Queequeg"}. This record does not assert independent reproduction. ### SeResNeXt101-STD Accuracy: 95.41. Result ID: 699aab4c5698665ec4384ffc. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"N/A"}. This record does not assert independent reproduction. ### DENSER Accuracy: 95.3. Result ID: 699aab4c5698665ec4384ff2. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@fillassuncao"}. This record does not assert independent reproduction. ### MobileNet Accuracy: 95. Result ID: 699aab4c5698665ec4384ff3. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@Bojone"}. This record does not assert independent reproduction. ### CNN with optional shortcuts Accuracy: 94.7. Result ID: 699aab4c5698665ec4384ff4. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@kennivich"}. This record does not assert independent reproduction. ### Google AutoML Accuracy: 93.9. Result ID: 699aab4c5698665ec4384ff5. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@Sebastian Heinz"}. This record does not assert independent reproduction. ### BitLogic Accuracy: 93.81. Result ID: 6998e1f75698665ec4463361. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"FFN","Gate count":"384 K"}. This record does not assert independent reproduction. ### Capsule Network Accuracy: 93.6. Result ID: 699aab4c5698665ec4384ff6. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@XifengGuo"}. This record does not assert independent reproduction. ### VGG16 Accuracy: 93.5. Result ID: 699aab4c5698665ec4384ff7. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@QuantumLiu"}. This record does not assert independent reproduction. ### LeNet Accuracy: 93.4. Result ID: 699aab4c5698665ec4384ff8. [Reporting paper]() Recorded evaluation configuration: {"Submitter":"@cmasch"}. This record does not assert independent reproduction. ### DWN (n=2) Accuracy: 89.12. Result ID: 6998e1f75698665ec446335f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"FFN","Gate count":null}. This record does not assert independent reproduction. ### DWN (n=6) Accuracy: 89.01. Result ID: 6998e1f75698665ec4463360. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"FFN","Gate count":null}. This record does not assert independent reproduction. ### BitLogic Accuracy: 81.07. Result ID: 6998e1f75698665ec4463362. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"CNN","Gate count":"253.4 K"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.