README.md

---
license: mit
base_model:
- Ultralytics/YOLO11
tags:
- ComputerVision
- Yolo
- Tomatoes
---
# TomatoACLv1

## Model Overview

**TomatoACLv1** is an object detection model trained from scratch using the **YOLO11n** architecture for tomato detection in images.

This repository includes the trained model weights in both **PyTorch (`.pt`)** and **ONNX (`.onnx`)** formats, along with training configuration and result files.

## Dataset

The model was trained using the following Kaggle dataset:

**Tomato Detection**  
https://www.kaggle.com/datasets/andrewmvd/tomato-detection

## Training Details

- Architecture: **YOLO11n**
- Training strategy: **trained from scratch**
- Task: **object detection**
- Target object: **tomato**

## Repository Contents

- `tomatoACLv1.pt` — trained PyTorch model
- `tomatoACLv1.onnx` — exported ONNX model
- `args.yaml` — training arguments
- `data.yaml` — dataset configuration
- `metrics.json` — training metrics
- `results.png` — training summary image

## Training Results

Below is the training summary image generated during training:

![Training results](./results.png)

## Intended Use

This model is intended for tomato detection in images and can be used for inference in environments compatible with:

- **PyTorch**
- **ONNX Runtime**
- other ONNX-compatible deployment frameworks

## Limitations

Performance may vary depending on image quality, lighting conditions, occlusions, and differences between real-world data and the original training dataset distribution.