set y axis to start from 0 in matplotlib

Question

I have a simple bar chart, which looks like:

The code that produced it is:

import matplotlib.pyplot as plt
import numpy as np

a = np.array([['c++','php'],
     [3,9]])
y_pos = np.arange(len(a[0]))
labels = np.arange(1,11,1)
objects = a[1]
ax = plt.bar(a[0], a[1], align='center', alpha=1)
plt.xticks(y_pos, a[0])
plt.ylabel('usage')
plt.title('Programming Languange Usage')

plt.show()

O made this bar chart, using matplotlib, the data is being taken from a 2x2 matrix, the problem is, the yaxis start from 3 and stop at 9, I want it to start from 0 and stop at 10. The problem that causing this is because I set the yticks from the matrix it self, thus it gave me this number.

Thee question is, how do I set the yticks to start from 0 and stop at 10 ?, I tried using np.arange() , and pyplot.yticks.

Does this answer your question? [setting y-axis limit in matplotlib](https://stackoverflow.com/questions/3777861/setting-y-axis-limit-in-matplotlib) — luigigi, May 07 '20 at 05:51
@luigigi That answers didnt work beacuse the array was of type `dtype=' — David, May 07 '20 at 06:17
You can convert the y values as you're plotting: `bars = plt.bar(a[0], a[1].astype(int))` — DavidG, May 07 '20 at 08:07

David · Accepted Answer · 2020-05-07T06:18:38.490

Changing the a from numpy array to list of list will solve the issue:

import matplotlib.pyplot as plt
import numpy as np

a = [['c++','php'],
     [3,9]]
y_pos = np.arange(len(a[0]))
labels = np.arange(1,11,1)
objects = a[1]
bars = plt.bar(a[0], a[1], align='center', alpha=1)
plt.xticks(y_pos, a[0])
plt.ylabel('usage')
plt.title('Programming Languange Usage')

plt.show()

This will give the following:

The problem is that the type of the values in the array is dtype='<U3' meaning that it will plot a char value in the y axis.

You can always separate the array to the x labels and y values and use numpy.

set y axis to start from 0 in matplotlib

1 Answers1