Matplotlib：如何用不同线型、标记与颜色区分绘制插值数据段

Hey there! Great question—your initial idea of plotting the full interpolated curve plus the interpolated segments with distinct styles is totally feasible, and we can make it work by first tracking which values were interpolated. Let's break this down step by step.

Step 1: Track Which Values Were Interpolated #

First, we need to keep a record of where the original data had missing values (since those are exactly the points that got interpolated). We'll create a boolean mask from the original DataFrame before interpolation.

import pandas as pd
import matplotlib.pyplot as plt

# Load your original data
data = {
    'datetime': ['2012-10-02 10:00:00', '2012-10-02 11:00:00', '2012-10-02 12:00:00',
                 '2012-10-02 13:00:00', '2012-10-02 14:00:00', '2012-10-02 15:00:00',
                 '2012-10-02 16:00:00', '2012-10-02 17:00:00', '2012-10-02 18:00:00',
                 '2012-10-02 19:00:00', '2012-10-02 20:00:00', '2012-10-02 21:00:00',
                 '2012-10-02 22:00:00', '2012-10-02 23:00:00'],
    'Phoenix': [30.0, 30.0, 31.0, 31.0, 32.0, 17.0, 28.0, None, 9.0, None, 13.0, None, 7.0, 6.0],
    'Chicago': [63.0, 63.0, 62.0, 62.0, 62.0, 51.0, None, None, 42.0, None, None, None, 48.0, 48.0]
}

df = pd.DataFrame(data)
df['datetime'] = pd.to_datetime(df['datetime'])
df.set_index('datetime', inplace=True)

# Create a mask to track original missing values
original_missing = df[['Phoenix', 'Chicago']].isna()

# Perform interpolation
df_interpolated = df.interpolate()

Step 2: Plot Full Curve + Highlight Interpolated Segments #

Now we'll first plot the complete interpolated curve with a standard style, then overlay the interpolated segments with a distinct format (like dashed lines and hollow markers) to make them stand out while keeping the overall curve coherent.

# Plot the full interpolated curves with solid lines
plt.plot(df_interpolated.index, df_interpolated['Phoenix'], color='#1f77b4', linestyle='-', label='Phoenix')
plt.plot(df_interpolated.index, df_interpolated['Chicago'], color='#ff7f0e', linestyle='-', label='Chicago')

# Highlight interpolated segments for Phoenix
phoenix_interp_x = df_interpolated.index[original_missing['Phoenix']]
phoenix_interp_y = df_interpolated['Phoenix'][original_missing['Phoenix']]
plt.plot(phoenix_interp_x, phoenix_interp_y, color='#1f77b4', linestyle='--', marker='o', markerfacecolor='white', markersize=8, label='Phoenix (Interpolated)')

# Highlight interpolated segments for Chicago
chicago_interp_x = df_interpolated.index[original_missing['Chicago']]
chicago_interp_y = df_interpolated['Chicago'][original_missing['Chicago']]
plt.plot(chicago_interp_x, chicago_interp_y, color='#ff7f0e', linestyle='--', marker='s', markerfacecolor='white', markersize=8, label='Chicago (Interpolated)')

# Add plot formatting for readability
plt.legend()
plt.xticks(rotation=45)
plt.tight_layout()
plt.ylabel('Temperature')
plt.show()

Simplified Alternative: Highlight Only Interpolated Points #

If you don't need to style the entire interpolated segment (just the individual points), using plt.scatter() is a simpler, cleaner approach:

# Plot full curves first
plt.plot(df_interpolated.index, df_interpolated['Phoenix'], color='#1f77b4', linestyle='-', label='Phoenix')
plt.plot(df_interpolated.index, df_interpolated['Chicago'], color='#ff7f0e', linestyle='-', label='Chicago')

# Add scatter points for interpolated values
plt.scatter(phoenix_interp_x, phoenix_interp_y, color='#1f77b4', marker='o', edgecolor='#1f77b4', facecolor='white', s=80, label='Phoenix Interpolated')
plt.scatter(chicago_interp_x, chicago_interp_y, color='#ff7f0e', marker='s', edgecolor='#ff7f0e', facecolor='white', s=80, label='Chicago Interpolated')

plt.legend()
plt.xticks(rotation=45)
plt.tight_layout()
plt.ylabel('Temperature')
plt.show()

Why This Works #

By preserving the original missing value mask, we can precisely identify which points in the interpolated DataFrame were generated by interpolation. Plotting these points/segments with a different style creates a clear visual distinction while maintaining the overall continuity of the curve—exactly what you were aiming for.

内容的提问来源于stack exchange，提问作者Samuel