kendrickfff
/

YOLOv8GarbageClassification

@@ -1,171 +1,138 @@
----
-license: mit
----
-Waste Classification - YOLOv8 Large (HighRes)
-🗑️ Model Summary
-This model is a fine-tuned version of YOLOv8 Large (yolov8l) designed for the detection and classification of 12 different types of waste materials. It was trained on high-resolution images (1280px) to capture fine details in trash objects.
-This project was developed by Kendrick as a Capstone Mini-Project for the REA AI Engineering Bootcamp.
-Model Architecture: YOLOv8-Large
-Task: Object Detection
-Input Resolution: 1280x1280
-Classes: 12
-🎯 Intended Use
-This model is intended to be used for:
-Automated waste sorting systems.
-Educational apps for recycling.
-Environmental monitoring.
-📊 Evaluation Results
-The model was evaluated on a validation set of 1,935 images.
-Metric
-Score
-Description
-mAP@50
-0.783
-Strong overall performance across all classes.
-Precision
-0.797
-High confidence in correct predictions.
-Recall
-0.733
-Good ability to find objects, though struggles with organic waste.
-mAP@50-95
-0.576
-High localization accuracy (bounding boxes are tight).
-Performance by Class
-The model excels at distinct rigid objects but struggles with amorphous organic matter.
-Class
-mAP@50
-Status
-Insight
-Clothes
-0.987
-🌟 Excellent
-Distinct shapes make this the easiest class.
-Brown-Glass
-0.905
-✅ Very Good
-Consistent texture and shape.
-Shoes
-0.847
-✅ Good
-High recall and precision.
-Plastic
-0.706
-⚠️ Moderate
-Struggles with transparency and deformation.
-Biological
-0.580
-❌ Needs Work
-High false negative rate.
-🔍 Deep Dive: Evaluation & Limitations
-Based on the Confusion Matrix analysis, the model demonstrates specific behaviors:
-The "Biological" Challenge:
-The model has a low Recall (0.445) for the biological class.
-Issue: It missed 741 instances of biological waste, misclassifying them as "Background".
-Reason: Biological waste (food scraps, leaves) lacks a defined shape compared to bottles or cans, blending into the background.
-False Positives:
-There are 540 instances where the model predicted biological waste on empty backgrounds. This suggests potential noise in the training labels or confusion with complex background textures.
-Background Confusion:
-The cardboard class also shows confusion with the background (159 false positives), likely due to color similarity with flooring/ground.
-⚙️ Training Configuration
-The model was trained on an NVIDIA A100-SXM4-40GB GPU for 6.26 hours.
-Epochs: 50
-Batch Size: 8
-Optimizer: AdamW (lr=0.000625)
-Image Size: 1280 (High Res)
-Augmentation: Standard YOLOv8 augmentations + Mosaic (disabled in last 10 epochs).
-💻 How to Use
-You can use this model with the Ultralytics Python library:
 from ultralytics import YOLO
-# Load the model
 model = YOLO("path/to/best.pt")
-# Perform inference on an image
 results = model("path/to/image.jpg")
 # Display results
 results[0].show()
-👨‍💻 Author
-Kendrick
-Alumni, REA AI Engineering Bootcamp
-This model card was generated based on training logs and evaluation charts.

+# 🗑️ Waste Classification – YOLOv8 Large (High Resolution)
+A high-resolution waste-classification model built using **YOLOv8-Large**, fine-tuned to detect and categorize **12 types of waste**.
+This model was developed as a Capstone Mini-Project for the **REA AI Engineering Bootcamp**.
+---
+## 📌 Overview
+This project provides a custom object-detection model designed to support:
+* ♻️ **Automated waste-sorting systems**
+* 📱 **Recycling education applications**
+* 🌍 **Environmental monitoring tools**
+The model was trained on 1280×1280 high-resolution images to better capture fine-grained details common in trash and recyclables.
+---
+## 🧠 Model Details
+| Property              | Value                  |
+| --------------------- | ---------------------- |
+| **Architecture**      | YOLOv8-Large (yolov8l) |
+| **Task**              | Object Detection       |
+| **Input Size**        | 1280 × 1280            |
+| **Number of Classes** | 12                     |
+| **Base Model**        | `ultralytics/yolov8l`  |
+| **License**           | MIT                    |
+---
+## 🎯 Classes
+The model detects 12 waste categories, including:
+* Clothes
+* Brown-Glass
+* Shoes
+* Plastic
+* Biological
+* (and others)
+---
+## 📊 Evaluation Results
+The model was evaluated on **1,935 validation images**.
+### **Overall Performance**
+| Metric     | Score     | Description                          |
+| ---------- | --------- | ------------------------------------ |
+| **mAP@50** | **0.783** | Strong overall detection performance |
+---
+### **Performance by Class**
+The model performs extremely well on rigid, well-shaped objects but struggles with amorphous organic materials.
+| Class           | mAP@50 | Status              | Insight                         |
+| --------------- | ------ | ------------------- | ------------------------------- |
+| **Clothes**     | 0.987  | 🌟 Excellent        | Consistent shape and texture    |
+| **Brown-Glass** | 0.905  | ✅ Very Good         | Strong geometric patterns       |
+| **Shoes**       | 0.847  | ✅ Good              | High recall and precision       |
+| **Plastic**     | 0.706  | ⚠️ Moderate         | Transparency/deformation issues |
+| **Biological**  | 0.580  | ❌ Needs Improvement | Blends into background          |
+---
+## 🔍 Deep Dive: Key Insights
+### **1. Biological Waste Challenge**
+* **Recall:** 0.445
+* **Missed detections:** 741 biological items labeled as background
+* **Cause:** Organic waste lacks distinct shape or edges, making it harder for YOLO to detect.
+### **2. False Positives**
+* **540 biological false positives** on plain backgrounds
+* Possibly caused by:
+  * Noisy labels
+  * Complex textures that resemble organic material
+### **3. Background Confusion**
+* Cardboard items have **159 false positives** due to color similarity with ground surfaces.
+---
+## ⚙️ Training Configuration
+| Setting           | Value                                               |
+| ----------------- | --------------------------------------------------- |
+| **Hardware**      | NVIDIA A100-SXM4-40GB                               |
+| **Training Time** | 6.26 hours                                          |
+| **Epochs**        | 50                                                  |
+| **Batch Size**    | 8                                                   |
+| **Optimizer**     | AdamW (lr = 0.000625)                               |
+| **Image Size**    | 1280                                                |
+| **Augmentations** | Standard YOLOv8 + Mosaic (disabled final 10 epochs) |
+---
+## 💻 Usage
+Install Ultralytics:
+```bash
+pip install ultralytics
+```
+Run inference:
+```python
 from ultralytics import YOLO
+# Load model
 model = YOLO("path/to/best.pt")
+# Run inference
 results = model("path/to/image.jpg")
 # Display results
 results[0].show()
+```
+---
+## 👨‍💻 Author
+**Kendrick**
+Alumni — REA AI Engineering Bootcamp
+This README and model card were generated with training logs and evaluation outputs.