Installation
Install PyRFDT and RFDT-Core to get started with RF digital twin simulation
Prerequisites
System Requirements
Software
- • Python: 3.8 or higher (3.10 recommended)
- • pip: Latest version
- • Git: For source installation
- • Conda: Anaconda or Miniconda
Hardware & Platform
- • GPU: NVIDIA GPU (CUDA-capable)
- • Windows: Windows 10/11 (64-bit)
- • Linux: Ubuntu 20.04+ or equivalent
Step 0: Create Conda Environment
First, create a dedicated conda environment with the required dependencies:
Create and activate conda environment
conda create --name RFDT_demo -y python=3.8
conda activate RFDT_demoInstall requirements
conda install -c conda-forge gcc=12.1.0
python -m pip install --upgrade pip
pip install torch==2.1.2+cu118 torchvision==0.16.2+cu118 --extra-index-url https://download.pytorch.org/whl/cu118
pip install numpy matplotlib opencv-python ipykernel
conda install -c "nvidia/label/cuda-11.8.0" cuda-toolkitStep 1: Install PyRFDT
PyRFDT is the Python package that provides the API and server for RF digital twin simulation.
Install from GitHub
Install the latest version directly from the GitHub repository:
pip install git+https://github.com/RFDigitalTwin/PyRFDTStep 2: Install RFDT-Core
RFDT-Core is the high-performance C++ ray tracing engine. Install the pre-built wheel for your platform:
Windows Installation
Python 3.8Pre-built wheel for Windows (Python 3.8):
pip install https://github.com/RFDigitalTwin/PyRFDT/releases/download/v0.01a/rfdt_rt-0.0.1-cp38-cp38-win_amd64.whlPlatform-specific wheels
RFDT-Core is a compiled C++ library with CUDA support. Make sure to install the correct wheel for your platform and Python version.
Verify Installation
Verify that both packages are installed correctly:
Test Import
python -c "import rfdt; print('PyRFDT version:', rfdt.__version__)"
python -c "import invertwin; print('RFDT-Core imported successfully')"
# Or in Python:
import rfdt
import invertwin as wt
print("✓ All packages installed successfully!")
print(f"PyRFDT: {rfdt.__version__}")
print(f"InverTwin (RFDT-Core): Available")