This page is desktop-only. It does not cover web service deployment.
- Python: 3.10+ recommended (follow project requirements)
- Core dependencies:
- PySide6 (GUI)
- mpmath (multiprecision computation)
- matplotlib (plots)
- Pillow (images and PDF preview conversion)
If you want PDF compilation from the desktop app:
- Install a TeX engine such as
pdflatex/xelatex(TeX Live / MiKTeX) - Ensure the desktop app can invoke the executable (PATH or a configured engine path)
Use an isolated environment (Conda or venv):
python -m venv .venv
source .venv/bin/activate
pip install -r gui_requirements.txt
python data_extrapolation_gui.pyThis repository provides PyInstaller configs/scripts (see the files in the repo):
DataLab.specbuild_mac_data_gui.shbuild_windows_data_gui.ps1/build_windows_data_gui.bat
- Bundle resources
The desktop app loads resources at runtime (icons and offline docs). Make sure the build includes:
docs/desktop/(offline desktop documentation)- Icon assets (
DataLab.pngor.icns/.ico)
- Temporary directory permissions
The app generates LaTeX/PDF intermediate files and preview images in a temporary directory:
- Windows: ensure
%TEMP%is writable - macOS/Linux: ensure the system temp directory is writable
- Fonts and non-ASCII text
If you need CJK typesetting or non-ASCII UI rendering:
- Prefer
xelatexand install appropriate fonts - Verify fonts are available on the target machine
- Windows: use
build_windows_data_gui.ps1; consider VC++ runtime and Qt plugins - macOS: use
build_mac_data_gui.sh; consider signing/notarization andMACOSX_DEPLOYMENT_TARGET - Linux: build and test on the target distro; verify Qt plugins and system libs
- Display formatting:
- Scientific OFF: decimal places
- Scientific ON: significant digits
- Log axes in fitting plots:
log-x/log-ywith automatic fallback - Export: CSV, LaTeX, PDF (depends on TeX engine availability)