=======
Copyright (2015-2018) C. Le Losq.
Rampy is a Python library that aims at helping processing spectroscopic data, such as Raman, Infrared or XAS spectra. It offers, for instance, functions to subtract baselines as well as to stack, resample or smooth spectra. It aims at facilitating the use of Python in processing spectroscopic data. It integrates within a workflow that uses Numpy/Scipy as well as optimisation libraries such as lmfit or emcee, for instance.
The /examples/ folder contain various examples.
Rampy is tested on Python 2.7 and 3.6 (see Travis badge; no garantee that it works on other Python versions)
The following libraries are required and indicated in setup.py:
- Scipy
- Numpy >= 1.12
- sklearn
- pandas
Optional dependencies:
- gcvspline (you need a working FORTRAN compiler for its installation. To avoid this problem under Windows, wheels for Python 2.7, 3.4 and 3.6 are provided for 64 bit Windows, and a wheel for Python 3.6 is provided for Windows 32 bits. If installation fails, please check if is due to a fortran compiler issue.)
Installation of gcvspline is necessary for use of the rampy.rameau() class.
- cvxpy v 1.0 or higher. As for gcvspline, the installation of cvxpy can cause problems for Windows users due to missing compiler. See instructions from cvxpy in this case.
Installation of gcvspline is necessary for use of the rampy.mixing() class.
Additional libraries for model fitting may be wanted:
- lmfit & aeval (http://cars9.uchicago.edu/software/python/lmfit/)
- emcee
Install with pip:
pip install rampy
If you want to use gcvspline and cvxpy, also install it:
pip install gcvspline
pip install cvxpy
Given a signal [x y] containing a peak, and recorded in a text file myspectrum.txt.
You can import it, remove a automatic background, plot the result, and print the centroid of the peak as:
import matplotlib.pyplot as plt
import numpy as np
import rampy as rp
spectrum = np.genfromtxt("myspectrum.txt")
bir = np.array([[0,100., 200., 1000]]) # the frequency regions devoid of signal, used by rp.baseline()
y_corrected, background = rp.baseline(spectrum[:,0],spectrum[:,1],bir,"arPLS",lam=10**10)
plt.figure()
plt.plot(spectrum[:,0],spectrum[:,1],"k",label="raw data")
plt.plot(spectrum[:,0],background,"k",label="background")
plt.plot(spectrum[:,0],y_corrected,"k",label="corrected signal")
plt.show()
print("Signal centroid is %.2f" % rp.centroid(spectrum[:,0],y_corrected))
See the /example folder for further examples.
Updated September 2018