CODEWITHSUNDEEP
pythongraphmatplotlib
Rahul
Rahul

Reputation: 11679

Overlapping bars in horizontal Bar graph

I am generating horizontal bar graphs using mathplotlib from
SO: How to plot multiple horizontal bars in one chart with matplotlib. The problem is when I have more than 2 horizontal bars, the bars are getting overlapped. Not sure, what I am doing wrong. Here is the following graph code

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

df = pandas.DataFrame(dict(graph=['q1','q2','q3' , 'q4','q5',' q6'],
                           n=[3, 5, 2,3 ,5 , 2], m=[6, 1, 3, 6 , 1 , 3])) 

ind = np.arange(len(df))
width = 0.4
opacity = 0.4
fig, ax = plt.subplots()


ax.barh(ind, df.n, width, alpha=opacity, color='r', label='Existing')
ax.barh(ind + width, df.m, width, alpha=opacity,color='b', label='Community')
ax.barh(ind + 2* width, df.m, width, alpha=opacity,color='g', label='Robust')

ax.set(yticks=ind + width , yticklabels=df.graph, ylim=[2*width - 1, len(df)])
ax.legend()



#plt.xlabel('Queries')
plt.xlabel('Precesion')
plt.title('Precesion for these  queries')

plt.show()

Currently, the graph looks like this Overlapped bars in Graph

Upvotes: 0

Views: 3516

Answers (2)

karthik reddy
karthik reddy

Reputation: 11

ax=df.plot.barh(width=0.85,figsize=(15,15))

adjust width(should be less than 1 otherwise overlaps) and figsize to get the best and clear view of bars. Because if the figure is bigger you can have a clear and bigger view of bars which is the ultimate goal.

Upvotes: 1

BrenBarn
BrenBarn

Reputation: 251598

You set the width of the bars to 0.4, but you have three bars in each group. That means the width of each group is 1.2. But you set the ticks only 1 unit apart, so your bars don't fit into the spaces.

Since you are using pandas, you don't really need to do all that. Just do df.plot(kind='barh') and you will get a horizontal bar chart of the dataframe data. You can tweak the display colors, etc., by using various paramters to plot that you can find in the documentation. (If you want the "graph" column to be used as y-axis labels, set it as the index: df.set_index('graph').plot(kind='barh'))

(Using df.plot will give a barplot with only two bars per group, since your DataFrame has only two numeric columns. In your example, you plotted column m twice, which doesn't seem very useful. If you really want to do that, you could add a duplicate column into the DataFrame.)

Upvotes: 2

Related Questions