README.md

---
library_name: onnx
tags:
  - efficientnet
  - image-classification
  - imagenet
  - computer-vision
  - onnx
  - inference4j
license: apache-2.0
pipeline_tag: image-classification
---

# EfficientNet-Lite4 — ONNX

ONNX export of [EfficientNet-Lite4](https://huggingface.co/onnx/EfficientNet-Lite4), a lightweight and efficient image classification model optimized for mobile/edge deployment. Trained on ImageNet with 1000-class output.

Mirrored for use with [inference4j](https://github.com/inference4j/inference4j), an inference-only AI library for Java.

## Original Source

- **Repository:** [ONNX](https://huggingface.co/onnx/EfficientNet-Lite4)
- **License:** apache-2.0

## Usage with inference4j

```java
try (EfficientNet model = EfficientNet.fromPretrained("models/efficientnet-lite4")) {
    List<Classification> results = model.classify(Path.of("cat.jpg"));
    results.forEach(c -> System.out.printf("%s: %.2f%%%n", c.label(), c.score() * 100));
}
```

## Model Details

| Property | Value |
|----------|-------|
| Architecture | EfficientNet-Lite4 (compound-scaled CNN) |
| Task | Image classification (ImageNet 1000 classes) |
| Input | `[batch, 224, 224, 3]` — RGB, pixel values 0-255 |
| Output | `[batch, 1000]` — class probabilities |
| ONNX opset | 11 |
| Original framework | TensorFlow Lite → ONNX |

## License

This model is licensed under the [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). Original model from [ONNX](https://huggingface.co/onnx).