README.md

---
license: mit
task_categories:
- object-detection
language:
- en
pretty_name: Cassava-Dataset (YOLOv8 Format)
size_categories:
- 10K<n<100K
configs:
- config_name: default
  data_files: []
---


# 🌿 Cassava-Dataset (YOLOv8 Format)

## 📌 Overview

The **Cassava-Dataset** is a computer vision dataset designed for **plant disease detection in cassava leaves**.  
It is formatted for **YOLOv8 object detection models**, making it suitable for training deep learning systems for agricultural disease identification.

This dataset was prepared for research and development in **smart agriculture, IoT farming systems, and AI-based plant disease detection**.

---

## 🎯 Objective

To enable automated detection and classification of cassava leaf diseases using deep learning models such as:

- YOLOv8 Nano (yolov8n)
- YOLOv8 Small/Medium
- Other object detection architectures

---

## 🧠 Classes

The dataset contains the following classes:

names:
- CBB
- CBSD
- CGM
- CMD
- HEALTHY

---

## 📁 Dataset Structure
cassava-7/
├── train/
│ ├── images/
│ └── labels/
├── valid/
│ ├── images/
│ └── labels/
├── test/
│ ├── images/
│ └── labels/
└── data.yaml


---

## ⚙️ YOLOv8 Configuration

Example `data.yaml`:

```yaml
path: .
train: train/images
val: valid/images
test: test/images

names:
 names:
- CBB
- CBSD
- CGM
- CMD
- HEALTHY
```

## 🚀 Usage (YOLOv8 Training)
Install dependencies
```yaml
pip install ultralytics

from ultralytics import YOLO

model = YOLO("yolov8n.pt")

model.train(
    data="data.yaml",
    epochs=50,
    imgsz=640,
    batch=16,
    device=0
)

```