Plotting Two Data Frame Value Counts Problems with the X-Axis Labels

0
votes

I plot value counts for two different data frames, but the second data frame is automatically plotted with the wrong x-axis labels. (See the Photo) Is there a way to change the order of the bars in each plot or to make all plots on the figure align by the x-axis labels? Right now, they are only ordering in a descending order, but I need them to order based on the x-axis labels.

In the "df" data frame, their are 158 Females and 140 Males. In my the "signed" data frame, there are 13 Females and 19 Males. When I plot these, the plot shows 140/13 Females and 158/19 Males, when it should show 158/13 females and 140/19 Males. I have been able to use the "ascending" parameter on the value_counts method to reverse the order of the value counts, but this solution will not work for my data frames with more than two different values.

Resulting plot: plot

> df['Gender'].value_counts()

F    158
M    140
Name: Gender, dtype: int64

> signed['Gender'].value_counts()

M    19
F    13
Name: Gender, dtype: int64

df['Gender'].value_counts().plot('bar', position=1, width = width)
signed['Gender'].value_counts().plot('bar', color='skyblue', position=0, width = width)
width = .25
plt.show()

(see photo of plot above)

2
pythonpandasplotbar-chart
Welcome to SO. Please take the time to read How to Ask and the other links found on that page. There may be more than one way. Dataframe plots are a 'simple wrapper' for Matplotlib.pyplot.plt. You can typically find a feature you want by browsing Matplotlib's gallery It also has excellent Tutorials. which are worth investing time with. - wwii

2 Answers

2
votes

Consider consolidating into a single dataframe by joining two series on Gender index:

import matplotlib
matplotlib.style.use('ggplot')

...
graph_df = df['Gender'].value_counts().rename('df').to_frame()\
               .join(signed['Gender'].value_counts().rename('signed').to_frame())

graph_df.plot(kind='bar',figsize=(8, 4))

Plot Output

1
votes

Use sort_index() after value_counts()

width = .25    
df['Gender'].value_counts().sort_index().plot('bar', position=1, width = width)
signed['Gender'].value_counts().sort_index().plot('bar', color='skyblue', position=0, width = width)
plt.show()