Matplotliib: Creating multi bar charts for multiple columns from a Pandas dataframe

Question

I have a Pandas dataframe with four columns; the first is a centretype (x axis) and I want the rest of the columns to be displayed as side by side columns for each centretype by year. I am not sure what I'm doing wrong, as I am getting only the values for 2019 but I want 2018 and 2017 displayed too.

df = c2

colors = ['steelblue','forestgreen','salmon']

ax = df.plot(kind='bar', color=colors, width=0.7, align='center', stacked=False, rot=60, figsize=(12,6), legend=False, zorder=3)


from matplotlib.ticker import FuncFormatter, MaxNLocator
ax.yaxis.set_major_locator(MaxNLocator(integer=True))

plt.grid(zorder=0)
#plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.2))
plt.xlabel("Type of centre")
plt.ylabel("Number")
plt.title("")

ax.spines['bottom'].set_color('gainsboro')
ax.spines['top'].set_color('white') 
ax.spines['right'].set_color('white')
ax.spines['left'].set_color('gainsboro')

ax.yaxis.label.set_color('black')
ax.xaxis.label.set_color('black')

ax.title.set_color('black')

ax.patch.set_facecolor('white')

plt.savefig('images/figure1.png', dpi=300, facecolor=ax.get_facecolor(), transparent=True,  bbox_inches='tight', pad_inches=0.1)
plt.show()
data.shape

If I define df like df = pd.DataFrame(np.random.rand(5,3), columns=[2018, 2019, 2020], index=list("ABCDE")) and run your code I get the desired output, namely this. This means that you're either not using the data you think you do, or that you're using a specific pandas version with a bug in it. — ImportanceOfBeingErnest
– ImportanceOfBeingErnest, Commented Jan 20, 2020 at 17:14

Valdi_Bo · Accepted Answer · 2020-01-20 18:26:32Z

The reason that your picture contains just one column is df = c2. Apparently c2 contains only one of columns from your DataFrame. Delete this instruction.

As a test, I performed just:

df.plot.bar(width=0.7, rot=60, figsize=(12,6), legend=False);

getting the following proper picture, with 5 sets of bars, 3 bars each:

Start from just the above instruction and then decide which remaining instruction should actually be executed.

In my opinion, the omitted parameters are not needed, because they pass their respective default values.

Note also the semicolon at the end. Working under Jupyter Notebook it is required to render just the picture. Otherwise (without the terminating semicolon), <matplotlib.axes._subplots.AxesSubplot at ...> is printed over the picture.

Maybe you should also drop legend=False (stay with default True value and print the legend). Otherwise it is not clear what particular bars refer to.

Collectives™ on Stack Overflow

Matplotliib: Creating multi bar charts for multiple columns from a Pandas dataframe

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related