README.md

---
license: mit
datasets:
- garythung/trashnet
- Zesky665/TACO
- detection-datasets/coco
language:
- en
tags:
- object-detection
- computer-vision
- yolov5
---
# Examples 
<div align="center">
  <img width="416" alt="turhancan97/yolov5-detect-trash-classification" src="https://huggingface.co/turhancan97/yolov5-detect-trash-classification/resolve/main/example1.jpg">
</div>
<div align="center">
  <img width="416" alt="turhancan97/yolov5-detect-trash-classification" src="https://huggingface.co/turhancan97/yolov5-detect-trash-classification/resolve/main/example2.jpg">
</div>
<div align="center">
  <img width="416" alt="turhancan97/yolov5-detect-trash-classification" src="https://huggingface.co/turhancan97/yolov5-detect-trash-classification/resolve/main/example3.jpg">
</div>

### How to use

- Install [yolov5](https://github.com/fcakyon/yolov5-pip):

```bash
pip install -U yolov5
```

- Load model and perform prediction:

```python
import yolov5

# load model
model = yolov5.load('turhancan97/yolov5-detect-trash-classification')
  
# set model parameters
model.conf = 0.25  # NMS confidence threshold
model.iou = 0.45  # NMS IoU threshold
model.agnostic = False  # NMS class-agnostic
model.multi_label = False  # NMS multiple labels per box
model.max_det = 1000  # maximum number of detections per image

# set image
img = 'https://github.com/ultralytics/yolov5/raw/master/data/images/zidane.jpg'

# perform inference
results = model(img, size=416)

# inference with test time augmentation
results = model(img, augment=True)

# parse results
predictions = results.pred[0]
boxes = predictions[:, :4] # x1, y1, x2, y2
scores = predictions[:, 4]
categories = predictions[:, 5]

# show detection bounding boxes on image
results.show()

# save results into "results/" folder
results.save(save_dir='results/')
```

- Finetune the model on your custom dataset:

```bash
yolov5 train --data data.yaml --img 416 --batch 16 --weights turhancan97/yolov5-detect-trash-classification --epochs 10
```