Skip to content

PyTorch ,ONNX and TensorRT implementation of YOLOv4

License

Notifications You must be signed in to change notification settings

marvis/pytorch-YOLOv4

 
 

Repository files navigation

Pytorch-YOLOv4

A minimal PyTorch implementation of YOLOv4.

├── README.md
├── dataset.py            dataset
├── demo.py               demo to run pytorch --> tool/darknet2pytorch
├── demo_darknet2onnx.py  tool to convert into onnx --> tool/darknet2pytorch
├── demo_pytorch2onnx.py  tool to convert into onnx
├── models.py             model for pytorch
├── train.py              train models.py
├── cfg.py                cfg.py for train
├── cfg                   cfg --> darknet2pytorch
├── data            
├── weight                --> darknet2pytorch
├── tool
│   ├── camera.py           a demo camera
│   ├── coco_annotatin.py       coco dataset generator
│   ├── config.py
│   ├── darknet2pytorch.py
│   ├── region_loss.py
│   ├── utils.py
│   └── yolo_layer.py

image

0. Weights Download

0.1 darkent

0.2 pytorch

you can use darknet2pytorch to convert it yourself, or download my converted model.

1. Train

use yolov4 to train your own data

  1. Download weight

  2. Transform data

    For coco dataset,you can use tool/coco_annotatin.py.

    # train.txt
    image_path1 x1,y1,x2,y2,id x1,y1,x2,y2,id x1,y1,x2,y2,id ...
    image_path2 x1,y1,x2,y2,id x1,y1,x2,y2,id x1,y1,x2,y2,id ...
    ...
    ...
    
  3. Train

    you can set parameters in cfg.py.

     python train.py -g [GPU_ID] -dir [Dataset direction] ...
    

2. Inference (Evolving)

  • Image input size for inference

    Image input size is NOT restricted in 320 * 320, 416 * 416, 512 * 512 and 608 * 608. You can adjust your input sizes for a different input ratio, for example: 320 * 608. Larger input size could help detect smaller targets, but may be slower and GPU memory exhausting.

    height = 320 + 96 * n, n in {0, 1, 2, 3, ...}
    width  = 320 + 96 * m, m in {0, 1, 2, 3, ...}
  • Different inference options

    • Load the pretrained darknet model and darknet weights to do the inference (image size is configured in cfg file already)

      python demo.py -cfgfile <cfgFile> -weightfile <weightFile> -imgfile <imgFile>
    • Load pytorch weights (pth file) to do the inference

      python models.py <num_classes> <weightfile> <imgfile> <IN_IMAGE_H> <IN_IMAGE_W> <namefile(optional)>
    • Load converted ONNX file to do inference (See section 3 and 4)

    • Load converted TensorRT engine file to do inference (See section 5)

  • Inference output

    Inference output is of shape [batch, num_boxes, 4 + num_classes] in which [batch, num_boxes, 4] is x_center, y_center, width, height of bounding boxes, and [batch, num_boxes, num_classes] is confidences of bounding box for all classes.

    Until now, still a small piece of post-processing including NMS is required. We are trying to minimize time and complexity of post-processing.

3. Darknet2ONNX (Evolving)

  • This script is to convert the official pretrained darknet model into ONNX

  • Pytorch version Recommended: 1.4.0

  • Install onnxruntime

    pip install onnxruntime
  • Run python script to generate ONNX model and run the demo

    python demo_darknet2onnx.py <cfgFile> <weightFile> <imageFile> <batchSize>

    This script will generate 2 ONNX models.

    • One is for running the demo (batch_size=1)
    • The other one is what you want to generate (batch_size=batchSize)

4. Pytorch2ONNX (Evolving)

  • You can convert your trained pytorch model into ONNX using this script

  • Pytorch version Recommended: 1.4.0

  • Install onnxruntime

    pip install onnxruntime
  • Run python script to generate ONNX model and run the demo

    python demo_pytorch2onnx.py <weight_file> <image_path> <batch_size> <n_classes> <IN_IMAGE_H> <IN_IMAGE_W>

    For example:

    python demo_pytorch2onnx.py yolov4.pth dog.jpg 8 80 416 416

    This script will generate 2 ONNX models.

    • One is for running the demo (batch_size=1)
    • The other one is what you want to generate (batch_size=batch_size)

5. ONNX2TensorRT (Evolving)

  • TensorRT version Recommended: 7.0, 7.1

  • Run the following command to convert VOLOv4 ONNX model into TensorRT engine

    trtexec --onnx=<onnx_file> --explicitBatch --saveEngine=<tensorRT_engine_file> --workspace=<size_in_megabytes> --fp16
    • Note: If you want to use int8 mode in conversion, extra int8 calibration is needed.
  • Run the demo

    python demo_trt.py <tensorRT_engine_file> <input_image> <input_H> <input_W>
    • This demo here only works when batchSize=1, but you can update this demo a little for batched inputs.

    • Note1: input_H and input_W should agree with the input size in the original ONNX file.

    • Note2: extra NMS operations are needed for the tensorRT output. This demo uses python NMS code from tool/utils.py.

    • Inference on X86 is verified to be okay for TensorRT 7.0, but output of the first iteration each time engine is loaded may be wrong on Jetson platforms. If you are using Jetpack 4.4 DP on Jetson platforms, try to ignore the first iteration each time as a workaround.

6. ONNX2Tensorflow

Reference:

@article{yolov4,
  title={YOLOv4: YOLOv4: Optimal Speed and Accuracy of Object Detection},
  author={Alexey Bochkovskiy, Chien-Yao Wang, Hong-Yuan Mark Liao},
  journal = {arXiv},
  year={2020}
}

About

PyTorch ,ONNX and TensorRT implementation of YOLOv4

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%