# Unsupervised Image Classification on ImageNet 200 randomly selected classes (val)

Source: [Unsupervised Image Classification on ImageNet 200 randomly selected classes (val)](<https://www.sota2.com/research/sota/unsupervised-image-classification-on-imagenet-200-randomly-selected-classes-val>)

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

Data updated: 2026-02-26T06:06:36.851Z

Evaluation of unsupervised image classification performance on a specific subset of 200 randomly selected classes from the ImageNet validation set, utilizing clustering-related metrics such as NMI and ARI.

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 | Unsupervised image classification |
| Dataset (including any stated split) | ImageNet 200 randomly selected classes (val) |
| Metric | Top-1 Acc |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Top-1 Acc](<https://www.sota2.com/research/sota/unsupervised-image-classification-on-imagenet-200-randomly-selected-classes-val.md?metric=Top-1%20Acc>): Higher is better; unit: PERCENT.

- [Top-5 Acc](<https://www.sota2.com/research/sota/unsupervised-image-classification-on-imagenet-200-randomly-selected-classes-val.md?metric=Top-5%20Acc>): Higher is better; unit: PERCENT.

- [NMI](<https://www.sota2.com/research/sota/unsupervised-image-classification-on-imagenet-200-randomly-selected-classes-val.md?metric=NMI>): Higher is better; unit: PERCENT.

- [ARI](<https://www.sota2.com/research/sota/unsupervised-image-classification-on-imagenet-200-randomly-selected-classes-val.md?metric=ARI>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Acc | Result date | Report paper |
| --- | --- | --- | --- |
| SCAN | 58.1 | Not recorded | 68fa67d21e209131cea18ae5 |
| SCAN | 56.3 | Not recorded | 68fa67d21e209131cea18ae5 |
| K-means | 52.5 | Not recorded | 68fa67d21e209131cea18ae5 |

### SCAN

Top-1 Acc: 58.1. Result ID: 698d48378b517eb433205403.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18ae5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18ae5.md>)

[Implementation](<https://www.sota2.com/here>)

Recorded evaluation configuration: {"Stage":"Self-labeling step (†)"}. This record does not assert independent reproduction.

### SCAN

Top-1 Acc: 56.3. Result ID: 698d48378b517eb433205402.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18ae5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18ae5.md>)

[Implementation](<https://www.sota2.com/here>)

Recorded evaluation configuration: {"Stage":"Clustering step (\*)"}. This record does not assert independent reproduction.

### K-means

Top-1 Acc: 52.5. Result ID: 698d48378b517eb433205401.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67d21e209131cea18ae5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61641e209131ce9f408c.md>)

Recorded evaluation configuration: {"Pretext features":"MoCo \[8\]"}. This record does not assert independent reproduction.

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