Student: We learnt C, C++, Fortran, Matlab, Labview, Excell (sic)...
Why should we learn python? What is python?
Reply: because
I do not use Python for a long time but I can say:
During your study, Python is free
After your study, Python is .... free
Python works with any OS
Python is interpreted (I loved and still love QuickBasic)
The community is very dynamic and is made of fine scientists
Python allows data analysis and the graphics are beautiful
You can interface a DAQ, an instrument, a camera (keywords: pyvisa, pydaq, opencv)
Packages are easy to install thanks to anaconda
Seduced? What to do? Here is a procedure that will allow you to live on the free part of the world. Some help in French: https://www.youtube.com/watch?v=22qI92HJTq0
You can also directly install Anaconda https://www.continuum.io/downloads.
Download and install VirtualBox: https://www.virtualbox.org/wiki/Downloads
Download «.iso» Debian: https://www.debian.org/distrib/
Put «.iso» in a folder
Config the Virtual machine:
Start VirtualBox
Create a new Virtual Machine
Boot disk priority: CD/DVD then Hard disk
Change memory, processor and virtual disk as you want
Activate 3D for display
Stockage: Master IDE —> put the path to .iso
5. Start VM
6. Chose “graphic install” and “domain.local” and “Debian+XFCe” and install GRUB on the hard disk when you ask.
Start VM, it should work.
Open a terminal:
su
apt-get install python-idle
apt-get install python-matplotlib
apt-get install python-scipy
(option) apt-get install python-pip
C’est fait!
Example of a Boud’code
In a directory, I have a set of data files that which names have the same structure: «data.txt».
The following code allows to automatically analyse all the file «*.txt» contained in the directory defined by the variable path.
from pylab import *
from scipy.optimize import curve_fit
from scipy import interpolate
import os
import glob
# Curve fit procedures
def cur_pow(x,y):
def f(x,a,b):
return a*x**b
params = curve_fit(f, x, y)
[a,b]=params[0]
print params
xmax=max(x)
xmin=min(x)
x_new=np.linspace(xmin,xmax,1000)
plot(x_new,a*x_new**b,'-r')
return 0
def cur_exp(x,y,a0,b0):
def f(x,a,b):
return a*np.exp(-b*x)
guess=(a0,b0)
params = curve_fit(f, x, y,guess)
[a,b]=params[0]
xmax=max(x)
xmin=min(x)
x_new=np.linspace(xmin,xmax,1000)
print a,b
print params
plot(x_new,a*np.exp(-b*x_new),'g-')
return 0
#Positionning of the data
path='C:\Users\Administrator\Documents\Python Scripts'
os.chdir(path)
path2="*.txt"
dataf=glob.glob(path2) #dataf is a list containing all the names of the files.
# LOOP on ALL .txt files contained in the path
for d in dataf:
#read the data + initialisation of vectors
data=loadtxt(d,delimiter=',')
x=data[:,0]
y=data[:,1]
xaxis='time(s)'
yaxis='angle'
fig=figure(1)
pl=fig.add_subplot(2,1,1)
plot(x,y,'o')
pl=fig.add_subplot(2,1,2)
plot(y,x,'o')
show()
figure(2)
plot(x,x*x,'o')
cur_pow(x,x*x)
cur_exp(x,x*x,1,0.2)
show()