Installation¶
curvefit-stats is a lightweight package built on NumPy and SciPy. No pandas,
plotting library or notebook runtime is required.
Supported runtime¶
| Component | Supported range |
|---|---|
| Python | >=3.11; Linux compatibility tested on 3.11–3.14 |
| NumPy | >=1.26.4,<3 |
| SciPy | >=1.12,<1.19 |
The tested numerical stacks are recorded in release verification. Version ranges are compatibility policy, not an exhaustive test of every combination.
ODR compatibility
scipy.odr is deprecated from SciPy 1.17 and removed in 1.19. EIV calls emit
an explicit FutureWarning on deprecated backends. The dependency upper
bound excludes removal; no replacement backend is currently configured.
Install the source checkout¶
Until a registry release is published, use the repository:
git clone https://github.com/bibymaths/curvefit-stats.git
cd curvefit-stats
uv venv --python 3.12
uv pip install --python .venv/bin/python .
.venv/bin/python -c "import curvefit_stats; print(curvefit_stats.__all__)"
git clone https://github.com/bibymaths/curvefit-stats.git
cd curvefit-stats
python -m venv .venv
python -m pip --python .venv/bin/python install .
.venv/bin/python -c "import curvefit_stats; print(curvefit_stats.__all__)"
On Windows, replace .venv/bin/python with .venv\Scripts\python.exe.
An editable development checkout can use uv pip install -e . after activating
its environment. Development instructions define the separate,
locked quality-check environment; installing the base package installs no dev tools.
Install built artifacts¶
For a wheel or source distribution you have built locally:
uv pip install --python .venv/bin/python dist/curvefit_stats-0.1.0-py3-none-any.whl
# Or build/install from the source distribution:
uv pip install --python .venv/bin/python dist/curvefit_stats-0.1.0.tar.gz
See packaging checks before sharing artifacts. Proceed to the quickstart once imports succeed.