https://github.com/Iywie/pl_YOLO

YOLOv7, YOLOX and YOLOv5 are working right now
https://github.com/Iywie/pl_YOLO

object-detection pytorch pytorch-lightning yolo yolov5 yolov7 yolox

Last synced: 12 days ago
JSON representation

YOLOv7, YOLOX and YOLOv5 are working right now

• Host: GitHub
• URL: https://github.com/Iywie/pl_YOLO
• Owner: Iywie
• Created: 2022-01-21T10:45:55.000Z (over 2 years ago)
• Default Branch: master
• Last Pushed: 2023-02-04T03:54:49.000Z (over 1 year ago)
• Last Synced: 2024-02-29T07:32:34.245Z (4 months ago)
• Topics: object-detection, pytorch, pytorch-lightning, yolo, yolov5, yolov7, yolox
• Language: Python
• Homepage:
• Size: 158 MB
• Stars: 25
• Watchers: 1
• Forks: 6
• Open Issues: 4
• Metadata Files:
  • Readme: README.md

Lists

• awesome-yolo-object-detection - Iywie/pl_YOLO

README

        
# pl-YOLO

Detection on pytorch lightning.

## Step of training

### 1. Build the model parameters and dataset parameters in configs.

### 2. Run

```python

  python train.py -c  -d 

```

## Example

### COCO2017 Dataset

#### Dataset directory

```

COCO2017  

└─annotations_trainval2017   

│   └─annotations   

│      │ instances_train2017.json   

│      │ instances_train2017.json   

└─train2017

└─val2017 

```

#### Change dataset parameters

In configs/data/coco2017.yaml, change the directory, image size and batch size.

```

    dir: 

    ...

    train_size: [640,640]

    val_size: [640,640]

    train_batch_size: 32

    val_batch_size: 32

```

#### Run YOLOX-s

```python

  python train.py -d configs/data/coco2017.yaml -c configs/model/yolox/yolox_s.yaml

```