Skip to content

User Guide

Purpose: Learn how to use FoodSpec for common tasks (loading data, running analyses, interpreting results).

Audience: Spectroscopy researchers and data scientists using FoodSpec in practice.

Time: 2–3 hours to read core sections; reference chapters as needed.

Prerequisites: Python basics; some familiarity with spectroscopy or food science.

What You'll Learn

This guide covers practical FoodSpec workflows: - Loading data: CSV, HDF5, vendor formats (OPUS/WiRE) - Preprocessing: Baseline correction, smoothing, normalization - Analysis: Classification, regression, mixture analysis - Quality control: Validation, outlier detection, reproducibility - Customization: Writing protocols, using plugins, extending FoodSpec

For quick start: See 15-minute quickstart
For theory: See theory section
For API details: See API reference

Example: Typical Workflow

from foodspec.io import load_csv
from foodspec.preprocess import baseline_als, normalize_snv
from foodspec.apps.oils import run_oil_authentication

# 1. Load (see: Data Formats & I/O)
ds = load_csv("oils.csv", wavenumber_col="wavenumber")

# 2. Preprocess (see: Preprocessing & Feature Extraction)
ds = baseline_als(ds, lam=1e6)
ds = normalize_snv(ds)

# 3. Analyze (see: Analysis Workflows)
results = run_oil_authentication(ds, label_column="oil_type")
print(f"Accuracy: {results.metrics['accuracy']:.2%}")

# 4. Validate (see: Validation & Reproducibility)
print(f"95% CI: [{results.metrics['accuracy_ci_lower']:.2%}, {results.metrics['accuracy_ci_upper']:.2%}]")

Command-Line Interface

Data Management

Input/Output

  • Data Formats & HDF5 — Supported file formats and HDF5 libraries
  • Vendor I/O — Import data from instrument vendors (Thermo, Bruker, etc.)
  • CSV to Library — Convert CSV files to FoodSpec libraries

Libraries & Organization

Configuration & Automation

Extensibility

Use this when: You need comprehensive documentation for a specific feature or workflow.