RPDNet is the model that can detect valid cells from millimeter wave radar RDM.

• Python 3.6
• PyTorch 1.8+
• GPU

• (Not necessary) Install Anaconda3
• Install CUDA 11.0
• install cuDNN8.0
• Install PyTorch

Noted that our code is tested based on PyTorch 1.8

Download Dataset.

Before training or test, please make sure you have prepared the dataset by organizing the directory as: data/your_dataset/data and data/your_dataset/label. E.g. data/GT6/data and data/GT6/label.

In config/base_confige.yml, you might want to change the following settings:

• data (NECESSARY) root path of the dataset for training or testing
• WORK_PATH path to save/load checkpoints
• CUDA_VISIBLE_DEVICES indices of GPUs
• learning_rate learning rate
• batch_size batch size for traning

Train a model by

python main.py train

• --config path of configuration file #Default: config/base_config.yml

Evaluate the trained model by

python main.py eval

• --config path of configuration file #Default: config/base_config.yml
• --epoch iteration of the checkpoint to load. #Default: -1

It will output Precision rate, Recall rate, and number of target points

Transform the prepared dataset using the trained model by

python main.py transform

• --config path of configuration file #Default: config/base_config.yml
• --epoch iteration of the checkpoint to load. #Default: -1

RDNetis freely available for free non-commercial use, and may be redistributed under these conditions.