How to plot two DataFrame on same graph for comparison

python pandas matplotlib graph data-analysis
Zythyr picture Zythyr · Jul 7, 2015 · Viewed 20.3k times · Source

I have two DataFrames (trail1 and trail2) with the following columns: Genre, City, and Number Sold. Now I want to create a bar graph of both data sets for a side by side comparison of Genre vs. total Number Sold. For each genre, I want to two bars: one representing trail 1 and the other representing trail 2.

How can I achieve this using Pandas?

I tried the following approach which did NOT work. 

gf1 = df1.groupby(['Genre'])
gf2 = df2.groupby(['Genre']) 
gf1Plot = gf1.sum().unstack().plot(kind='bar, stacked=False)
gf2Plot = gf2.sum().unstack().plot(kind='bar, ax=gf1Plot, stacked=False)

I want to be able to see How trail1 data set compared to trial2 data for each of the Genre (ex: Spicy, Sweet, Sour, etc...)

I also tried using concat, but I can't figure out how to graph the concatenated DataFrame on the same graph to compare the two keys. 

DF = pd.concat([df1,df2],keys=['trail1','trail2'])

Answer

Zythyr picture Zythyr · Jul 7, 2015

I found a solution to my question. I welcome others to post a better approach.

Solution:

df1 = pd.DataFrame(myData1, columns=['Genre', 'City', 'Sold'])
df2 = pd.DataFrame(myData2, columns=['Genre', 'City', 'Sold'])

df1['Key'] = 'trail1'
df2['Key'] = 'trail2'

DF = pd.concat([df1,df2],keys=['trail1','trail2'])

DFGroup = DF.groupby(['Genre','Key'])

DFGPlot = DFGroup.sum().unstack('Key').plot(kind='bar')

Here is an example of the generated graph: enter image description here

Related questions

Plot correlation matrix using pandas

I have a data set with huge number of features, so analysing the correlation matrix has become very difficult. I want to plot a correlation matrix which we get using dataframe.corr() function from pandas library. Is there any built-in …

Read More
Add x and y labels to a pandas plot

Suppose I have the following code that plots something very simple using pandas: import pandas as pd values = [[1, 2], [2, 5]] df2 = pd.DataFrame(values, columns=['Type A', 'Type B'], index=['Index 1', 'Index 2']) df2.plot(lw=2, colormap='jet', marker='.', …

Read More
How to plot two columns of a pandas data frame using points?

I have a pandas data frame and would like to plot values from one column versus the values from another column. Fortunately, there is plot method associated with the data-frames that seems to do what I need: df.plot(x=…

Read More