Skip to main content
This is a pre-production deployment of Warehouse. Changes made here affect the production instance of PyPI (
Help us improve Python packaging - Donate today!

A python module using scipy's orthogonal distance regression that makes fitting data easy.

Project Description

The python-fit module is designed for people who need to fit data frequently and quickly. The module is not designed for huge amounts of control over the minimization process but rather tries to make fitting data simple and painless. If you want to fit data several times a day, every day, and you really just want to see if the fit you’ve made looks good against your data, check out this software. If you want one very statistically aware and neurotically controlled fit, you might consider looking elsewhere (give zunzun a look).

With python-fit, you get work done.


  • Common fitting curves built-in
  • Default parameters for built-in functions intelligently calculated using your data.
  • Fit with user defined functions, too.
  • A ready-to-plot-fit always conviently returned.
  • Get fit parameters and associated errors.
  • Chi-squared residual.

Example Usage

Fit with built in functions:

from pylab import *
import fit
from numpy import random, exp

# Create some data to fit
x = arange(-10, 10, .2)
# A gaussian of height 10, width 2, centered at zero. With noise.
y = 10*exp(-x**2/8) + (random.rand(100) - 0.5)

# No need to provide first guess at parameters for fit.gaus
(xf, yf), params, err, chi =, x,y)

print "N:    %.2f +/- %.3f" % (params[0], err[0])
print "N:    %.2f +/- %.3f" % (params[1], err[1])
print "N:    %.2f +/- %.3f" % (params[2], err[2])

plot(x,y, 'bo', label='Data')
plot(xf, yf, 'r-', label='Fit')

Functions available:

Gaussian curve
Exponential curve
Double exponential
Polynomials for degrees 0-20
Crystal Ball
... and more!

Fit with user defined functions:

x = (4.2105303, 5.2631601, 6.2405997, 7.5187997, 8.7218, 9.7744402, 10.676691, 11.65414, 12.63158, 13.83459, 14.887219, 16.015039, 17.06767,
18.270679, 19.24812, 20.300751, 21.50376, 23.157888, 25.789471, 28.345871, 30.601501, 33.458643, 39.022559, 46.015039, 48.270679)
y = (0.18942001, 0.2099, 0.23891001, 0.27816002, 0.31911, 0.35836001, 0.39932001, 0.43686003, 0.46416002, 0.49829001, 0.51536004, 0.52556, 0.51876995,
0.5, 0.47271, 0.44026, 0.39249001, 0.33106002, 0.24060, 0.17746, 0.13311001, 0.11262, 0.095566, 0.095566, 0.095566)

x = array(x)
y = array(y)

# Guassian with quadratic background.
def example_function(params, x):
    N,mu,sigma,a,b,c = params
    return N*exp(-0.5 * ((x-mu)/sigma)**2 ) + a*x**2 + b*x + c

# It will still try to guess parameters, but they are dumb!
(xf,yf),p,e,chi =, x,y)
plot(x,y, 'bo', label='Data')
plot(xf,yf, 'r-', label='Fit')

Even though example_function is defined by the user, python-fit will guess parameters (the median value of the xdata for all parameters; it works if x and y are on similar scales). If the fit fails, then provide some decent parameters as a first guess:

results =, x, y, default_pars = [1, 12, 10, 1, 1, 1])
plot(results[0][0], results[0][1], 'r--')

Fit a sub-range:

results =, x, y, data_range=[0, 23])
plot(results[0][0], results[0][1], 'r-.')

Define your own weights to prevent outliers from wreaking havoc on your fit:

# Create some outliers.
y_outlier = y + (random.rand(len(y))**20)*3
# I'll just make a cut and say outliers are above 0.55
weights = 1. * (y_outlier < .55)
results =, x, y_outlier, we=weights)
plot(x,y_outlier, 'bo', label='Data w/ Outliers')
plot(results[0][0], results[0][1], 'r-.', label='Fit around outliers')

Release History

This version
History Node


Download Files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, Size & Hash SHA256 Hash Help File Type Python Version Upload Date
(6.6 kB) Copy SHA256 Hash SHA256
Source None May 27, 2014

Supported By

Elastic Elastic Search Pingdom Pingdom Monitoring Dyn Dyn DNS Sentry Sentry Error Logging CloudAMQP CloudAMQP RabbitMQ Kabu Creative Kabu Creative UX & Design Google Google Cloud Servers Fastly Fastly CDN StatusPage StatusPage Statuspage DigiCert DigiCert EV Certificate