3D Geometry Dataset Generator

Overview

Authors: Igor Maurell (igormaurell@gmail.com), Pedro Corçaque (pedrollcorc@gmail.com)

The 3D Geometry Dataset Generator software uses Pythonocc and Gmsh API to process BRep CAD files to generate a machine learning geometry dataset.

Formats

Input Formats

• STEP, BRep, Iges (CAD File)
• YAML (Metadata File)

Output Formats

• JSON, YAML, PKL (Geometry Features)
• PLY (CAD Mesh)
• JSON (Stats)

Dependencies

Just a list of dependencies, follow the installation steps to install them all.

• Anaconda
• Pythonocc (>= 7.7.0)
• Gmsh API
• argparse
• tqdm
• PyYAML
• numpy

Installation

Apt dependencies:

sudo apt-get install software-properties-common cmake
sudo apt-get install libtool autoconf automake gfortran gdebi
sudo apt-get install gcc-multilib libxi-dev libxmu-dev libxmu-headers
sudo apt-get install libx11-dev mesa-common-dev libglu1-mesa-dev
sudo apt-get install libfontconfig1-dev
sudo apt-get install libfreetype6 libfreetype6-dev
sudo apt-get install tcl tcl-dev tk tk-dev
sudo apt-get install libfltk1.3-dev

Create Conda Env

Create conda env:

conda create --name=pyoccenv python=3.9
conda activate pyoccenv

OCC (7.7.0)

Download the "tgz" file on https://dev.opencascade.org/release, than:

cd ~/Downloads
tar xf opencascade-7.7.0.tgz
cd opencascade-7.7.0
mkdir build && cd build
cmake ..
make
sudo make install

Pythonocc (7.7.0)

conda install -c conda-forge pythonocc-core=7.7.0

3D Geometry Dataset Generator

On your workspace folder, download:

git clone https://github.com/AeroScan/3DGeometryDatasetGenerator
cd 3DGeometryDatasetGenerator
pip install -r requirements.txt

Using

After install, you can use:

conda activate pyoccenv
python data_generator.py --help