Skip to content

Linear regression

Project status: Complete

This work is part of my first Rust project (see here). The aim is to learn how to write Python bindings for Rust code, combining the high-performance Rust kernel and Python flexibility with enterprise-grade DevOps.

Roadmap

  • Write Python API wrapper for the Rust library crate.
  • Add CI pipeline for automatic publishing to PyPI.

Installation

python -m venv my-venv
source my-venv/bin/activate
pip install drkaizeng-linear-regression

uv venv my-venv
source my-venv/bin/activate
uv pip install drkaizeng-linear-regression

Usage

As a CLI tool

Assume that an environment with the package installed has been activated. 

linear-regression --help

As an importable library

from pathlib import Path

from linear_regression.linear_regression import linear_regression

input_tsv = Path("input.tsv")
with open(input_tsv, "w") as f:
    f.write("1\t2\n2\t3\n3\t4\n")

output_tsv = Path("output.tsv")

linear_regression.callback(input_tsv=input_tsv, output_tsv=output_tsv, overwrite=True)

Note that linear_regression is a function decorated with @click.command(), which turns it into a click.core.Command object. We need to use its .callback attribute to access the underlying function.