# Da Profile_class_test

%reset
%load_ext autoreload
%autoreload 2

Once deleted, variables cannot be recovered. Proceed (y/[n])? y

%qtconsole

%matplotlib inline

import matplotlib.pyplot as plt
import numpy as np
import os

from dataIO.span import span
from dataIO.fn_add_subfix import fn_add_subfix

from IPython.display import display
from plotting.backends import maximize

pwd

'C:\\Users\\kovor\\Documents\\python\\pyXTel\\pyxsurf\\pyProfile\\test'

Test pyProfile

New class implementation (2020/06/25)

Test new implementation of class from profile methods to objects, in analogy to what is done with pySurf. Here we test and document.

The main class is Profile. Little is implemented for readers, it can be easily initialized with x and y.

Helper function make_signal (see Appendix or make_signal? for details) can be used to generate a (sinusoid-based) test profile.

from pyProfile.profile_class import Profile
from pyProfile.profile import make_signal

Profile?

make_signal?

x,y=make_signal(amp=10.,L=30.,N=21,nwaves=2.8,ystartend=(0,0),noise=0)
plt.plot(x,y)

[<matplotlib.lines.Line2D at 0x1e3205ef8d0>]

Units can be set. Use Profile methods:

P = Profile(x,y,units=['mm','nm'])

P.std()

7.044127837632114

P.plot()

[<matplotlib.lines.Line2D at 0x1e32066b1d0>]

P = Profile(x,y,units=['mm','nm'])
P.plot(stats=1,loc=1)

[<matplotlib.lines.Line2D at 0x1e3206ee0f0>]

x and y can be retrieved as P.x and P.y, or with x,y = P()

P()

(array([ 0. ,  1.5,  3. ,  4.5,  6. ,  7.5,  9. , 10.5, 12. , 13.5, 15. ,
        16.5, 18. , 19.5, 21. , 22.5, 24. , 25.5, 27. , 28.5, 30. ]),
 array([10.        ,  6.3742399 , -1.87381315, -8.7630668 , -9.29776486,
        -3.09016994,  5.35826795,  9.92114701,  7.28968627, -0.6279052 ,
        -8.09016994, -9.68583161, -4.25779292,  4.25779292,  9.68583161,
         8.09016994,  0.6279052 , -7.28968627, -9.92114701, -5.35826795,
         3.09016994]))

#TODO: test remove_nan_ends.
#TODO: test register_profile.

Test profile functions

Make different test profiles.

BEWARE: units in algebraic operations are not verified, usually the ones from first term are used for result, this may change in future.

Create two similar quadratic profiles a and b with different x values:

# Make different test profiles:

x0 = np.arange(5)

a = Profile(x0,x0**2)
a.plot(marker='o',ls='-',label = 'a')

b =  Profile(x0+0.5,(x0+0.5)**2)
b.plot(marker='x',ls='-',label = 'b')

plt.legend(loc=0)

<matplotlib.legend.Legend at 0x1e3201ea470>

Algebraic operations can be performed on Profile objects.

Resampling can be directly accessed by resample method, but there is usually no need to perform, because it is automatically handled by algebraic operations (resample on first by default, ):

c = a.resample(b)

# plot interpolation

a.plot(marker='+',ls='-',label = 'a')
b.plot(marker='x',ls='-',label = 'b')
c.plot(marker='o',ls='',label='a resampled on b')

plt.legend(loc=0)

<matplotlib.legend.Legend at 0x1e3204c9160>

Here some examples of algebraic operations on different `x`. 

a.plot(marker='o',ls='-', label = 'a')
b.plot(marker='x',ls='-', label = 'b')
(a+b).plot(label = 'a + b',marker='o')
((a+b)/2).plot(label = '(a + b)/2',marker='d',ls='')
plt.grid()
plt.legend(loc=0)

<matplotlib.legend.Legend at 0x1e321a44fd0>

Appendix: make_signal

%matplotlib

Creo un segnale (reale). Ne creo una copia sottocampionata per test, ovviamente sono tutti valori reali.

L=1
N=11

plt.clf()
x,y=make_signal(1,L=1,N=101,nwaves=3)
plt.plot(x,y,label='harmonic signal with 0 phase')
x,y=make_signal(1,L=1,N=N,nwaves=3)
plt.plot(x,y,'o',label='downsampled curve')
x,y=make_signal(1,L=1,N=101,nwaves=3,phase=np.pi/4)
plt.plot(x,y,label='phase pi/4')
plt.grid(1)
plt.legend(loc=0)

<matplotlib.legend.Legend at 0x2a55959feb8>

Quindi ora posso creare segnali complessi arbitrari usando due segnali con stessa frequenza e fase diversa su reale e immaginario.

Questo ovviamente significa che posso usare una serie di ca N/2 armoniche per descrivere interamente il segnale, a patto di fornire anche la fase.

Option minus_one

plt.clf()
A=make_signal(1,L=1,N=N,nwaves=nwaves)
plt.plot(*A)
print(A[0].shape)
print(A[1].shape)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-27-c506e6c6b8ec> in <module>
      1 plt.clf()
----> 2 A=make_signal(1,L=1,N=N,nwaves=nwaves)
      3 plt.plot(*A)
      4 print(A[0].shape)
      5 print(A[1].shape)

NameError: name 'nwaves' is not defined

<Figure size 432x288 with 0 Axes>

# redo plot adding "minux_one option"
plt.clf()
plt.plot(*A)

plt.plot(*(make_signal(1,L=1,N=N,nwaves=nwaves,minus_one=True)),'o')

plt.grid(1)
plt.xlim([0.97,1.01])
plt.ylim([-0.2,0.1])
plt.show()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-28-6f760a9ed699> in <module>
      1 # redo plot adding "minux_one option"
      2 plt.clf()
----> 3 plt.plot(*A)
      4 
      5 plt.plot(*(make_signal(1,L=1,N=N,nwaves=nwaves,minus_one=True)),'o')

NameError: name 'A' is not defined

<Figure size 432x288 with 0 Axes>

# riproduce esempio di matlab da:
# https://it.mathworks.com/help/matlab/data_analysis/detrending-data.html
#
#

fn = r'C:\Users\kovor\Documents\python\pyXTel\pyProfile\test\matlab-normaldata.dat'

y = np.genfromtxt(fn)
p=Profile(np.arange(len(y)),y)
p.plot(color='b',label = 'raw')
(p-p.level()).plot(color='b',label = 'fit')
p.level().plot(color='r',label = 'leveled')
plt.plot(p.x,p.y*0+p.level().y.mean(),color='r',label = 'avg leveled')
plt.legend()

<matplotlib.legend.Legend at 0x2a5594a3588>

# makes a 2d surface from matlab data 
# fit 2D
# estrai profili dritti o in diagonale