---
license: apache-2.0
library_name: libreyolo
pipeline_tag: image-classification
datasets:
  - imagenet-1k
tags:
  - image-classification
  - efficientnetv2
  - libreyolo
---

# LibreEfficientNetV2b2-cls

EfficientNetV2-base-b2 image classifier (1000-class ImageNet-1k), repackaged
for [LibreYOLO](https://github.com/LibreYOLO/libreyolo). Eval resolution
260px; timm-reported top-1 accuracy ~80.5%.

## Usage

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreEfficientNetV2b2-cls.pt")
result = model.predict("image.jpg")[0]
print(result.probs.top1, result.probs.top1conf)
print(result.probs.top5)
```

## Source

Derived from the timm checkpoint
[`tf_efficientnetv2_b2.in1k`](https://huggingface.co/timm/tf_efficientnetv2_b2.in1k) in
[huggingface/pytorch-image-models](https://github.com/huggingface/pytorch-image-models).
Copyright (c) 2019 Ross Wightman. Licensed under the Apache License 2.0.

Original architecture: EfficientNetV2 by Google
([google/automl](https://github.com/google/automl/tree/master/efficientnetv2)),
"EfficientNetV2: Smaller Models and Faster Training"
([arXiv:2104.00298](https://arxiv.org/abs/2104.00298)), Apache License 2.0.
Only the ImageNet-1k checkpoint is published here — the ImageNet-21k / JFT
variants carry extra-data terms and are intentionally excluded.

## Modifications

State-dict key remapping only. Learned parameters are unchanged; inference is
bit-identical to timm (`max_abs_diff == 0`). See
`weights/convert_efficientnetv2_weights.py` in the
[LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo).

## License

Apache License 2.0. See the [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE)
files in this repository.