Pandas.plot（subplots = True）在每个子图中有3列

Question

I have a DataFrame with 700 rows and 6 columns:

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

df = pd.DataFrame(np.random.rand(700,6))

I can plot all columns in a single plot by calling:

df.plot()

And I can plot each column in a single plot by calling:

df.plot(subplots=True)

How can I have two subplots with three columns each from my DataFrame?!

Answer 1

Here's a general approach to plot a dataframe with n columns in each subplot:

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

df = pd.DataFrame(np.random.rand(700,6))

col_per_plot = 3
cols = df.columns.tolist()
# Create groups of 3 columns
cols_splits = [cols[i:i+col_per_plot] for i in range(0, len(cols), col_per_plot)]  

# Define plot grid.
# Here I assume it is always one row and many columns. You could fancier...
fig, axarr = plt.subplots(1, len(cols_splits))
# Plot each "slice" of the dataframe in a different subplot
for cc, ax in zip(cols_splits, axarr):
    df.loc[:, cc].plot(ax = ax)

This gives the following picture: