UnitModule

Installation

This project is based on MMDetection.

• Python 3.8
• Pytorch 1.11.0+cu113

Step 1. Create a conda virtual environment and activate it.

conda create -n unitmodule python=3.8 -y
conda activate unitmodule

Step 2. Install PyTorch following official instructions.

Linux and Windows

# Wheel CUDA 11.3
pip install torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio==0.11.0 --extra-index-url https://download.pytorch.org/whl/cu113

# Conda CUDA 11.3
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch

Step 3. Install MMDetection and dependent packages.

pip install -U openmim
mim install mmengine==0.7.4
mim install mmcv==2.0.0
mim install mmdet==3.0.0
mim install mmyolo==0.5.0
pip install -r requirements.txt

Dataset

The data structure DUO looks like below:

# DUO

data
├── DUO
│   ├── annotations
│   │   ├── instances_train.json
│   │   ├── instances_test.json
│   ├── images
│   │   ├── train
│   │   ├── test

Training

bash tools/dist_train.sh configs/yolox/yolox_s_100e_duo.py 2

Test

bash tools/dist_test.sh configs/yolox/yolox_s_100e_duo.py yolox_s_100e_duo.pth 2