# Image Classification on SVHN unlearned set

Source: [Image Classification on SVHN unlearned set](<https://www.sota2.com/research/sota/image-classification-on-svhn-unlearned-set>)

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

Data updated: 2026-04-23T09:40:31.385Z

Evaluation of Image Classification performance on the specific 'unlearned set' subset of the SVHN dataset, reporting Micro F1-score.

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) | SVHN unlearned set |
| Metric | Micro F1-score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Micro F1-score](<https://www.sota2.com/research/sota/image-classification-on-svhn-unlearned-set.md?metric=Micro%20F1-score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Micro F1-score | Result date | Report paper |
| --- | --- | --- | --- |
| L-CODEC | 95 | Not recorded | 68fa787f1e209131cea78b1e |
| Original | 94.53 | Not recorded | 68fa787f1e209131cea78b1e |
| Fisher | 94.23 | Not recorded | 68fa787f1e209131cea78b1e |
| Fine Tune | 93.73 | Not recorded | 68fa787f1e209131cea78b1e |
| Certified | 93.73 | Not recorded | 68fa787f1e209131cea78b1e |
| Retrain | 93.13 | Not recorded | 68fa787f1e209131cea78b1e |
| Neg Grad | 92.13 | Not recorded | 68fa787f1e209131cea78b1e |

### L-CODEC

Micro F1-score: 95. Result ID: 69e9e8ff47f34ce924d177aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374b3.md>)

[Implementation](<https://github.com/vsingh-group/LCODEC-deep-unlearning>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Original

Micro F1-score: 94.53. Result ID: 69e9e8ff47f34ce924d177a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Fisher

Micro F1-score: 94.23. Result ID: 69e9e8ff47f34ce924d177a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67551e209131cea158d5.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Fine Tune

Micro F1-score: 93.73. Result ID: 69e9e8ff47f34ce924d177a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67551e209131cea158d5.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Certified

Micro F1-score: 93.73. Result ID: 69e9e8ff47f34ce924d177ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Retrain

Micro F1-score: 93.13. Result ID: 69e9e8ff47f34ce924d177a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Neg Grad

Micro F1-score: 92.13. Result ID: 69e9e8ff47f34ce924d177a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa787f1e209131cea78b1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67551e209131cea158d5.md>)

Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction.

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