# Classification on STL10 Affine (test)

Source: [Classification on STL10 Affine (test)](<https://www.sota2.com/research/sota/classification-on-stl10-affine-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-24T07:40:44.604Z

Evaluation of image classification performance on the Affine-transformed test split of the STL-10 dataset, reporting accuracy on both original and augmented data.

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 | Classification |
| Dataset (including any stated split) | STL10 Affine (test) |
| Metric | Accuracy (STL10 Affine Original) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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 (STL10 Affine Original)](<https://www.sota2.com/research/sota/classification-on-stl10-affine-test.md?metric=Accuracy%20(STL10%20Affine%20Original)>): Higher is better; unit: PERCENT.

- [Accuracy (STL10 Affine Augmented)](<https://www.sota2.com/research/sota/classification-on-stl10-affine-test.md?metric=Accuracy%20(STL10%20Affine%20Augmented)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (STL10 Affine Original) | Result date | Report paper |
| --- | --- | --- | --- |
| MaWa | 98.65 | Not recorded | 68fa78971e209131cea792be |
| EC | 98.58 | Not recorded | 68fa78971e209131cea792be |
| WaCo | 98.42 | Not recorded | 68fa78971e209131cea792be |
| MaWa | 98.19 | Not recorded | 68fa78971e209131cea792be |
| WaCo | 98.19 | Not recorded | 68fa78971e209131cea792be |
| SCLR | 97.41 | Not recorded | 68fa78971e209131cea792be |
| SCLR | 97.4 | Not recorded | 68fa78971e209131cea792be |
| HSIC | 97.24 | Not recorded | 68fa78971e209131cea792be |
| HSIC | 96.73 | Not recorded | 68fa78971e209131cea792be |

### MaWa

Accuracy (STL10 Affine Original): 98.65. Result ID: 69eb1e7247f34ce924d307b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Nonlinear (NC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### EC

Accuracy (STL10 Affine Original): 98.58. Result ID: 69eb1e7247f34ce924d307b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"End-to-end (EC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### WaCo

Accuracy (STL10 Affine Original): 98.42. Result ID: 69eb1e7247f34ce924d307b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Nonlinear (NC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### MaWa

Accuracy (STL10 Affine Original): 98.19. Result ID: 69eb1e7247f34ce924d307ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Linear (LC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### WaCo

Accuracy (STL10 Affine Original): 98.19. Result ID: 69eb1e7247f34ce924d307ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Linear (LC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### SCLR

Accuracy (STL10 Affine Original): 97.41. Result ID: 69eb1e7247f34ce924d307b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Nonlinear (NC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### SCLR

Accuracy (STL10 Affine Original): 97.4. Result ID: 69eb1e7247f34ce924d307af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Linear (LC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### HSIC

Accuracy (STL10 Affine Original): 97.24. Result ID: 69eb1e7247f34ce924d307b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Nonlinear (NC)","Features":"DINOv2"}. This record does not assert independent reproduction.

### HSIC

Accuracy (STL10 Affine Original): 96.73. Result ID: 69eb1e7247f34ce924d307b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78971e209131cea792be.md>)

Recorded evaluation configuration: {"Classifier Type":"Linear (LC)","Features":"DINOv2"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.