# Image Classification on PneumoniaMNIST v2 (test)

Source: [Image Classification on PneumoniaMNIST v2 (test)](<https://www.sota2.com/research/sota/image-classification-on-pneumoniamnist-v2-test>)

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

Data updated: 2026-02-26T06:09:16.115Z

Evaluation of Image Classification performance on the test split of the PneumoniaMNIST v2 dataset, reporting g-mean, gamma\_RS, and number of GANs.

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) | PneumoniaMNIST v2 (test) |
| Metric | g-mean |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [g-mean](<https://www.sota2.com/research/sota/image-classification-on-pneumoniamnist-v2-test.md?metric=g-mean>): Higher is better; unit: PERCENT.

- [gamma\_RS](<https://www.sota2.com/research/sota/image-classification-on-pneumoniamnist-v2-test.md?metric=gamma_RS>): Higher is better; unit: SCALAR.

- [Num GANs](<https://www.sota2.com/research/sota/image-classification-on-pneumoniamnist-v2-test.md?metric=Num%20GANs>): Higher is better; unit: SCALAR.

## Recorded results

| Method | g-mean | Result date | Report paper |
| --- | --- | --- | --- |
| G\* (Oracle) | 88.1 | Not recorded | 68fa77af1e209131cea74b66 |
| G\* | 86.7 | Not recorded | 68fa77af1e209131cea74b66 |
| Single (top-1) | 85.4 | Not recorded | 68fa77af1e209131cea74b66 |
| Single (top-5) | 84.2 | Not recorded | 68fa77af1e209131cea74b66 |
| Naive (A) | 82.3 | Not recorded | 68fa77af1e209131cea74b66 |
| Real | 82.2 | Not recorded | 68fa77af1e209131cea74b66 |
| Naive (R) | 82.2 | Not recorded | 68fa77af1e209131cea74b66 |
| Single (average) | 65.2 | Not recorded | 68fa77af1e209131cea74b66 |

### G\* (Oracle)

g-mean: 88.1. Result ID: 699a470f5698665ec4ce587a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"G\* Oracle Selection","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### G\*

g-mean: 86.7. Result ID: 699a470f5698665ec4ce5879.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"G\* (Proposed Ensemble Selection)","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Single (top-1)

g-mean: 85.4. Result ID: 699a470f5698665ec4ce5874.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"top-1 ranking","Ranking source":"Real test set performance","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Single (top-5)

g-mean: 84.2. Result ID: 699a470f5698665ec4ce5875.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"top-5 ranking","Ranking source":"Real test set performance","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Naive (A)

g-mean: 82.3. Result ID: 699a470f5698665ec4ce5878.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"Naive All Ensemble","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Real

g-mean: 82.2. Result ID: 699a470f5698665ec4ce5873.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

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

### Naive (R)

g-mean: 82.2. Result ID: 699a470f5698665ec4ce5877.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"Naive Random Ensemble","Backbone":"ResNet18"}. This record does not assert independent reproduction.

### Single (average)

g-mean: 65.2. Result ID: 699a470f5698665ec4ce5876.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77af1e209131cea74b66.md>)

Recorded evaluation configuration: {"Selection strategy":"average of single GANs","Backbone":"ResNet18"}. This record does not assert independent reproduction.

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