Plot multiindex dataframe groupping internal level on one subplot

Question

I want to plot multiindex dataframe on one plot with sublots.

Dataframe example

import datetime
import random
base = datetime.datetime.today()
numdays = 7
date_list = [base - datetime.timedelta(days=x) for x in range(numdays)]
df_plot = pd.DataFrame({"y": random.choices(range(0, 100), k=16 * numdays)},
   index=pd.MultiIndex.from_product(
    [["A", "B", "C", "D"],
     ["m1", "m2", 'm3', 'm4'], 
     date_list],
    names=["Group", "metric", "date"]))
df_plot.head()

Group   metric  date     y
A   m1  2022-02-02      18
        2022-02-01      47
        2022-01-31      2

I want four subplots in one figure where each subplot corresponds to one group (A, B, C, D) and on each sublot there are four lines that correspond to each metric (m1, m2, m3, m4). X-axis is date, y-axis is cell value in one-column dataframe with multiindex (see above example).

I know that plotting may be done in several ways by groupping dataframe by index.

Is there elegant solution in one command like below? There must be four supplots with four lines on each subplot.

df_plot.plot(..., subplots=True, layout=(2, 2))

Quang Hoang · Accepted Answer · 2022-02-02 14:40:56Z

1

You can try reset the index before groupby, then manually plot into a pre-populated axis grid:

df_plot = df_plot['y'].unstack('metric').reset_index('Group')
fig, axes = plt.subplots(2,2)

for ax, (group, data) in zip(axes.ravel(), df_plot.groupby('Group')):
    data.plot(ax=ax)

Output:

answered Feb 2, 2022 at 14:40

Quang Hoang

151k11 gold badges64 silver badges86 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Plot multiindex dataframe groupping internal level on one subplot

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