kafe2 documentation

Welcome to kafe2, the Karlsruhe Fit Environment 2!

kafe2 is a data fitting framework designed for use in undergraduate physics lab courses. It provides a basic Python toolkit for fitting models to data as well as visualizing the data and the model function. It relies on Python packages such as numpy and matplotlib, and can use the Python interface to the minimizer Minuit contained in the data analysis framework ROOT or in the Python package iminuit.

• Installing kafe2
  • Requirements
  • Installation notes (Linux)
    • Optional: Install ROOT
    • Optional: Install iminuit
  • Installation notes (Windows)
    • Install iminuit
    • Install kafe2
    • Using kafe with ROOT under Windows
• Theoretical Foundation
  • Basic notions
    • Measurements and theoretical models
    • Parameter estimation
    • Choosing a cost function
    • Least-squares (“chi-squared”)
    • Negative log-likelihood
  • Types of datasets
  • Handling uncertainties
    • Gaussian uncertainties
    • Correlations
    • Other types of uncertainties
  • Cost functions
    • Least-squares (“chi-squared”) estimator
    • Negative log-likelihood estimator
• User Guide
  • Basic Fitting Procedure
    • Using kafe2go
    • Using Python
  • Example 1: Line Fit
    • kafe2go
    • Python
  • Example 2: Model Functions
    • kafe2go
    • Python
  • Example 3: Parameter Constraints
    • kafe2go
    • Python
  • Example 4: Non Linear Fits
    • kafe2go
    • Python
  • Example 5: Histogram Fit
    • kafe2go
    • Python
  • Example 6: Unbinned Fit
    • kafe2go
    • Python
• API documentation
  • kafe2 in a nutshell
  • Parameter estimation tools: fit
    • Indexed data
    • XY data
    • Histograms
    • Tools for fitting series of indexed measurements (indexed)
    • Tools for fitting xy data (xy)
    • Tools for fitting histograms (histogram)
    • Abstract base classes (_base)