Got it, let's tackle this problem. Your current code generates a 7x2 grid of subplots (one for each column in your DataFrame), and you want to merge every pair of adjacent subplots—meaning each row's two subplots become a single subplot that displays both of the corresponding series. Here are two straightforward approaches to achieve this:

Approach 1: Manual Subplot Creation (Simplest) #

Instead of letting pandas auto-generate subplots for every column, we'll create a grid of 7 single subplots (one per group of two columns) and plot both series from each group onto the same subplot.

import pandas as pd
import matplotlib.pyplot as plt

# Your original data preparation code
NG = [1,2,3,4,5,6,7]
dflist = []
l = [1,2,3,4,5,6]
b = [6,5,4,3,2,1]
for n in NG:
    df = pd.DataFrame(l)
    dflist.append(df)
    df2 = pd.DataFrame(b)
    dflist.append(df2)
df = pd.concat(dflist, axis = 1)

# Create a 7-row, 1-column grid of subplots
fig, axes = plt.subplots(nrows=7, ncols=1, figsize=(8, 14), sharey=True)

# Loop through each group of two columns and plot them on the same subplot
for group_idx in range(7):
    # Select the two columns for this group
    group_cols = df.iloc[:, [2*group_idx, 2*group_idx + 1]]
    # Plot both series on the current subplot
    group_cols.plot(grid=True, ax=axes[group_idx])
    # Add a title for clarity
    axes[group_idx].set_title(f'Group {group_idx + 1}')

# Adjust layout to prevent overlapping labels
plt.tight_layout()
plt.show()

Approach 2: Using GridSpec for Fine Layout Control #

If you need more control over the subplot layout (like adjusting spacing or merging specific grid cells), you can use matplotlib's GridSpec to explicitly merge the two columns in each row:

import pandas as pd
import matplotlib.pyplot as plt
from matplotlib.gridspec import GridSpec

# Your original data preparation code (same as above)
NG = [1,2,3,4,5,6,7]
dflist = []
l = [1,2,3,4,5,6]
b = [6,5,4,3,2,1]
for n in NG:
    df = pd.DataFrame(l)
    dflist.append(df)
    df2 = pd.DataFrame(b)
    dflist.append(df2)
df = pd.concat(dflist, axis = 1)

# Create a figure and a 7x2 grid specification
fig = plt.figure(figsize=(8, 14))
gs = GridSpec(7, 2, figure=fig)

# Loop through each row and merge the two columns into one subplot
for row_idx in range(7):
    # Merge the two columns in the current row into a single axis
    ax = fig.add_subplot(gs[row_idx, :])
    # Select and plot the two columns for this group
    group_cols = df.iloc[:, [2*row_idx, 2*row_idx + 1]]
    group_cols.plot(grid=True, ax=ax, sharey=True)
    ax.set_title(f'Group {row_idx + 1}')

plt.tight_layout()
plt.show()

Key Idea #

The core change here is moving away from subplots=True (which creates one subplot per column) to manually grouping your DataFrame columns and plotting each group on a single subplot. This gives you full control over which series share a subplot.

内容的提问来源于stack exchange，提问作者Artur Müller Romanov