Pest Detection Model - YOLO11
Introduction
This model was trained to detect various pests in agricultural settings using YOLO11. The goal of this model is to assist farmers and agronomists in identifying pests to help in better crop management. The model was trained on a custom dataset and has been optimized for accuracy and efficiency in identifying different types of pests.
Model Details
- Model File: best.pt
- Framework: YOLO11
- Dataset: https://www.kaggle.com/datasets/leonidkulyk/ip102-yolov5
Training Metrics
Below are the metrics and results from the training process:
| Epoch | Train Box Loss | Train Class Loss | Precision | Recall | mAP50 | mAP50-95 |
|---|---|---|---|---|---|---|
| 1 | 1.158 | 1.488 | 0.436 | 0.087 | 0.062 | 0.031 |
| 2 | 1.021 | 0.994 | 0.553 | 0.167 | 0.138 | 0.074 |
| 3 | 0.916 | 0.874 | 0.611 | 0.209 | 0.191 | 0.102 |
| 4 | 0.828 | 0.785 | 0.513 | 0.271 | 0.240 | 0.130 |
| 5 | 0.763 | 0.712 | 0.485 | 0.305 | 0.287 | 0.163 |
Note: Metrics include precision, recall, mAP at 50% (mAP50), and mAP across 50-95% (mAP50-95) confidence.
Additional Files
- Training Metrics CSV: A detailed CSV file containing training metrics and results.
Training Graphs
Below is the graph representing the model's training process, including metrics such as loss, precision, and recall.
