如何用Matplotlib在Python中实现视频同款图表？求Python 3代码

Got it! Looking at the 847s timestamp in the video you referenced, the chart is a grouped bar plot with error bars—perfect for comparing metrics across multiple categories and groups. Below is a complete Python 3 + Matplotlib implementation that matches that visual style closely, with comments to help you tweak it to your exact data:

Full Code Snippet #

import matplotlib.pyplot as plt
import numpy as np

# ----------------------
# Step 1: Prepare your data
# Replace these with your actual values from the video/your dataset
# ----------------------
category_names = ['Task A', 'Task B', 'Task C', 'Task D']
group1_averages = [82, 76, 91, 87]  # Example: Model X performance
group1_margins = [3.2, 4.1, 2.5, 3.0]  # Error (e.g., standard deviation/confidence interval)
group2_averages = [78, 84, 86, 89]  # Example: Model Y performance
group2_margins = [3.8, 3.3, 2.8, 2.2]

# ----------------------
# Step 2: Set up plot parameters
# ----------------------
bar_width = 0.35
x_coords = np.arange(len(category_names))  # Base positions for categories
fig, ax = plt.subplots(figsize=(10, 6))  # Set figure size

# ----------------------
# Step 3: Draw bars and error bars
# ----------------------
# First group of bars
bars_group1 = ax.bar(x_coords - bar_width/2, group1_averages, width=bar_width,
                     label='Model X', color='#1f77b4', alpha=0.8)
ax.errorbar(x_coords - bar_width/2, group1_averages, yerr=group1_margins,
            fmt='none', c='black', capsize=5)  # Error bars

# Second group of bars
bars_group2 = ax.bar(x_coords + bar_width/2, group2_averages, width=bar_width,
                     label='Model Y', color='#ff7f0e', alpha=0.8)
ax.errorbar(x_coords + bar_width/2, group2_averages, yerr=group2_margins,
            fmt='none', c='black', capsize=5)

# ----------------------
# Step 4: Add plot details (matches video's style)
# ----------------------
ax.set_xlabel('Task Type', fontsize=12)
ax.set_ylabel('Accuracy (%)', fontsize=12)
ax.set_title('Model Performance Comparison', fontsize=14, pad=20)
ax.set_xticks(x_coords)
ax.set_xticklabels(category_names, fontsize=10)
ax.legend(fontsize=11)

# Optional: Add value labels on top of each bar (like in the video)
def add_bar_labels(bars):
    for bar in bars:
        height = bar.get_height()
        ax.text(bar.get_x() + bar.get_width()/2., height,
                f'{height}%',
                ha='center', va='bottom', fontsize=10)

add_bar_labels(bars_group1)
add_bar_labels(bars_group2)

# Adjust layout to prevent label cutoff
plt.tight_layout()

# Show the plot
plt.show()

Customization Tips #

• More groups: If the video has 3+ groups, adjust bar_width (e.g., 0.25 for 3 groups) and add additional bar/errorbar calls with shifted x-coords (e.g., x_coords - bar_width, x_coords, x_coords + bar_width).
• Colors: Swap the color values to match the exact hues from the video (use a color picker tool if needed).
• Horizontal bars: Use ax.barh() instead of ax.bar() for a horizontal version—just adjust the x/y labels and coordinate logic accordingly.
• Error bar style: Tweak capsize to make the error bar caps wider/narrower, or change c to a different color for the error lines.

内容的提问来源于stack exchange，提问作者firnnauriel