mikful / thepaper

The paper

Some smart words, links, tables, figures and graphs

Quick Start Examples

The quickest way will be using colab notebooks to reproduce our results

- training

- inference

- all together

Install

Python>=3.7.0 is required with all requirements.txt installed including PyTorch>=1.7:

$ git clone https://github.com/leetoo/thepaper
$ cd thepaper
$ pip install -r requirements.txt

Inference

Inference with YOLOv5 based model you can execute agains your _ video / images _ here :

import torch 

# Model
model = ...  # or yolov5m, yolov5x, custom

# Images
img = 'https://... '  # or file, PIL, OpenCV, numpy, multiple

# Inference
results = model(img)

# Results
results.print()  # or .show(), .save(), .crop(), .pandas(), etc.

Inference with detect.py

detect.py runs inference on a variety of sources and saving results to runs/detect.

$ python detect.py --source 0  # webcam
                            file.jpg  # image 
                            file.mp4  # video
                            path/  # directory
                            path/*.jpg  # glob
                            'https://youtu.be/NUsoVlDFqZg'  # YouTube video
                            'rtsp://example.com/media.mp4'  # RTSP, RTMP, HTTP stream

Training

Run commands below to reproduce results on DataSetv5 dataset (dataset auto-downloads). Use the largest --batch-size your GPU allows (batch sizes shown for 16 GB devices).

$ python train.py --data $$$ --cfg $$$ --weights yolov5l --batch-size 64

Tutorials

• Train Custom Data  🚀 RECOMMENDED
• Tips for Best Training Results  ☘️ RECOMMENDED
• Weights & Biases Logging  🌟 NEW

todo ask do we need those tutorials ?

Environments and Integrations

Get started in seconds with our verified environments and integrations, including Weights & Biases for automatic YOLOv5 experiment logging. Click each icon below for details.

Contact

For issues running the paper please visit GitHub Issues. For business or professional support requests please visit https://www.fellowship.ai