Recording headers in text file and making plots with subsequent data

Question

I am having trouble parsing a text file that I created with another program. The text file looks something like this:

velocity 4
0 0
0.0800284750334461 0.0702333599787275
0.153911082737118 0.128537103048848
0.222539323234924 0.176328826156044
0.286621942300277 0.21464146333504
0.346732028739683 0.244229944930359
0.403339781262399 0.265638972071027
...
velocity 8
0 0
0.169153136373962 0.124121036173475
0.312016311613761 0.226778846267302
0.435889653693839 0.312371513797743
0.545354054604357 0.383832483710643
0.643486956562741 0.443203331839287
...

I want to grab the number in the same row as velocity (the header) and save it as the title of the plot of the subsequent data. Every other row apart from the header represents the x and y coordinates of a shooting ball.

So if I have five different headers, I would like to see five different lines on a single graph with a legend displaying the different velocities.

Here is my python code so far. I am close to what I want to get, but I am missing the first set of data (velocity = 4 m/s) and the colors on my legend don't match the line colors.

import matplotlib.pyplot as plt

xPoints = []
yPoints = []
fig, ax = plt.subplots()

with open('artilleryMotion.txt') as inf:

    for line in inf:
        column = line.split()

        if line.startswith("v"):
            velocity = column[1]
            ax.plot(xPoints, yPoints, label = '%s m/s' % velocity)
        else:
            xPoints.append(column[0])
            yPoints.append(column[1])

ax.legend()
plt.title("Ping-Pong Ball Artillery Motion")
plt.xlabel("distance")
plt.ylabel("height")
plt.ylim(ymin = 0)
ax.set_autoscaley_on(1)

I have been struggling with this for a while.

Edit_1: This is my output at the moment:

Artillery motion plot

Edit_2: I removed the indentation of the last lines of code. The color problem still occurs.

Edit_3: How would I go about saving the x and y points to a new array for each velocity? This may solve my issues.

Edit_4: Thanks to Charles Morris, I was able to create these plots. I just need to now determine if the initial upwards "arcing" motion by the ping pong ball for the higher velocities is representative of the physics or is a limitation of my code.

Artillery Motion Final

Answer 1

Edit: Ignore the old information, and see Solved solution below:

The following code works an example text file: input.txt

velocity 4
0 0
0.0800284750334461 0.0702333599787275
0.153911082737118 0.128537103048848
0.222539323234924 0.176328826156044
0.286621942300277 0.21464146333504
0.346732028739683 0.244229944930359
0.403339781262399 0.265638972071027
velocity 8
0 0
0.169153136373962 0.124121036173475
0.312016311613761 0.226778846267302
0.435889653693839 0.312371513797743
0.545354054604357 0.383832483710643
0.643486956562741 0.443203331839287

1) Import our text file

We use np.genfromtxt() for imports. In this case, we can Specify that dtype = float. This has the effect that the affect that Numbers are imported as 'Float' and thus, strings (in this case 'Velocity'), are imported as NaN.

Source: https://docs.scipy.org/doc/numpy/user/basics.io.genfromtxt.html How to use numpy.genfromtxt when first column is string and the remaining columns are numbers?

from matplotlib import pyplot as plt
from itertools import groupby
from numpy import NaN as nan

A = np.genfromtxt('input.txt',dtype=float)


>>>
array([[        nan,  4.        ],
       [ 0.        ,  0.        ],
       [ 0.08002848,  0.07023336],
       [ 0.15391108,  0.1285371 ],
       [ 0.22253932,  0.17632883],
       [ 0.28662194,  0.21464146],
       [ 0.34673203,  0.24422994],
       [ 0.40333978,  0.26563897],
       [        nan,  8.        ],
       [ 0.        ,  0.        ],
       [ 0.16915314,  0.12412104],
       [ 0.31201631,  0.22677885],
       [ 0.43588965,  0.31237151],
       [ 0.54535405,  0.38383248],
       [ 0.64348696,  0.44320333]])

2) Slice the imported array A

We can slice these arrays into separate X and Y arrays representing our X and Y values. Read up on array slicing in numpy here: https://docs.scipy.org/doc/numpy/reference/arrays.indexing.html

In this case, we take all values with index = 0 (X) and all values with index 1 (Y):

# x values
# y values   
X = A[:,0]
Y = A[:,1]    

>>> X = array([        nan,  0.        ,  0.08002848,  0.15391108,  0.22253932,
    0.28662194,  0.34673203,  0.40333978,         nan,  0.        ,
    0.16915314,  0.31201631,  0.43588965,  0.54535405,  0.64348696])

>>> Y = array([ 4.        ,  0.        ,  0.07023336,  0.1285371 ,  0.17632883,
    0.21464146,  0.24422994,  0.26563897,  8.        ,  0.        ,
    0.12412104,  0.22677885,  0.31237151,  0.38383248,  0.44320333])

3) Split the data for each velocity.

Here we desire to separate our X and Y values into those for each Velocity. Our X values are separated by Nan and our Y values are separated by 4,8,16....

Thus: For x, we split by nan. nan is a result of the genfromtxt() parsing Velocity as a float and returning nan.

Sources: numpy: split 1D array of chunks separated by nans into a list of the chunks Split array at value in numpy

For y, we split our array up on the numbers 4,8,16 etc. To do this, we exclude values that, when divided by 4, have zero remainder (using the % Python operator).

Sources: Split array at value in numpy How to check if a float value is a whole number Split NumPy array according to values in the array (a condition) Find the division remainder of a number How do I use Python's itertools.groupby()?

XX = [list(v) for k,v in groupby(X,np.isfinite) if k]
YY = [list(v) for k,v in groupby(Y,lambda x: x % 4 != 0 or x == 0) if k]


>>> 
XX = [[0.0,
0.080028475033446095,
0.15391108273711801,
0.22253932323492401,
0.28662194230027699
0.34673202873968301,
0.403339781262399],
[0.0,
0.16915313637396201,
0.31201631161376098,
0.43588965369383897,
0.54535405460435704,
0.64348695656274102]]

>>> YY =
[[0.0,
0.070233359978727497,
0.12853710304884799,
0.17632882615604401,
0.21464146333504,
0.24422994493035899,
0.26563897207102699],
[0.0,
0.124121036173475,
0.22677884626730199,
0.31237151379774297,
0.38383248371064299,
0.44320333183928701]]

4) Extract labels

Using a similar technique as above, we accept values = to our velocities 4,8,16 etc. In this case, we accept only those numbers which, when divided by 4, have 0 remainder, and are not 0. We then convert to a string and add m/s.

Ylabels = [list(v) for k,v in groupby(Y,lambda x: x % 4 == 0 and x != 0) if k]
Velocities = [str(i[0]) + ' m/s' for i in Ylabels]

>>> Y labels = [[4.0], [8.0]]
>>> Velocities = ['4.0 m/s', '8.0 m/s']

5) Plot

Plot values by index for each velocity.

fig, ax = plt.subplots()
for i in range(0,len(XX)):
    plt.plot(XX[i],YY[i],label = Velocities[i])
ax.legend()
plt.title("Ping-Pong Ball Artillery Motion")
plt.xlabel("distance")
plt.ylabel("height")
plt.ylim(ymin = 0)
ax.set_autoscaley_on(1) 

Code Altogether:

import numpy as np
from matplotlib import pyplot as plt
from itertools import groupby
from numpy import NaN as nan

A = np.genfromtxt('input.txt',dtype=float)

X = A[:,0]
Y = A[:,1]    

Ylabels = [list(v) for k,v in groupby(Y,lambda x: x % 4 == 0 and x != 0) if k]
Velocities = [str(i[0]) + ' m/s' for i in Ylabels]

XX = [list(v) for k,v in groupby(X,np.isfinite) if k]
YY = [list(v) for k,v in groupby(Y,lambda x: x % 4 != 0 or x == 0) if k]

fig, ax = plt.subplots()
for i in range(0,len(XX)):
    plt.plot(XX[i],YY[i],label = Velocities[i])
ax.legend()
plt.title("Ping-Pong Ball Artillery Motion")
plt.xlabel("distance")
plt.ylabel("height")
plt.ylim(ymin = 0)
ax.set_autoscaley_on(1)  

Old Answer:

The first time you iterate over all lines in the file, your xPoints and yPoints arrays are empty. Therefore, when you try and plot values for v = 4, you are plotting an empty array - hence your missing line.

You need to populate the arrays first, and then plot them. At the moment, you are plotting the values for v = 4 in the line labelled v = 8, and for v = 8, the values for v = 16 and so on.

Ignore: For the array population, try the following:

xPoints = []
yPoints = []
with open('artilleryMotion.txt') as inf:
    # initialize placeholder velocity variable
    velocity = 0
    for line in inf:
        column = line.split()

        if line.startswith("v"):
            velocity = column[1]

        else:
            xPoints.append({velocity: column[0]})
            yPoints.append({velocity: column[1]})

In the above, you save the data as a list of dictionaries (separate for x and y points), where the key is equal to the velocity that has been read in most recently, and the values are the x and y coordinates.

As a new velocity is read in, the placeholder variable velocity is updated and so the x and y values can be identified according the key that they have.

This allows you to Seaprate your plots by dictionary key (look up D.iteritems() D.items() ) and you can plot each set of points individually.