Machine Learning Academy · 课时

使用 Matplotlib 和 Seaborn 可视化数据

您将绘制直方图、散点图和相关性热力图，在建模前探索数据分布与关系

使用 Matplotlib 和 Seaborn 可视化数据 是 CoddyKit 上的免费 Machine Learning Academy 课时。这是第 4 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 Machine Learning Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Machine Learning Academy 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

Why Visualisation Matters in ML

Charts aren't just for reports — they're a diagnostic tool at every step. Matplotlib gives full control; Seaborn makes beautiful stats plots with less code.

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

# Set seaborn theme for all plots
sns.set_theme(style='whitegrid', palette='muted')

# Load a built-in dataset
df = sns.load_dataset('tips')
print(df.head())

Histograms: Understanding Distributions

A histogram bins a numeric column to show its shape — normal, skewed, or with outliers. Spotting skew tells you when a log transform might help your model.

import matplotlib.pyplot as plt
import seaborn as sns

df = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(12, 4))

# Raw distribution (right-skewed)
axes[0].hist(df['total_bill'], bins=30, edgecolor='white')
axes[0].set_title('Total Bill Distribution (Raw)')

# After log transform
import numpy as np
axes[1].hist(np.log(df['total_bill']), bins=30, edgecolor='white')
axes[1].set_title('Total Bill Distribution (Log Transformed)')

plt.tight_layout()
plt.show()

Scatter Plots: Feature Relationships

A scatter plot shows how two numbers relate — linear, curved, or full of outliers. Add color with hue to squeeze a third variable into the same view.

import matplotlib.pyplot as plt
import seaborn as sns

df = sns.load_dataset('tips')

# Scatter plot with hue encoding
sns.scatterplot(
    data=df,
    x='total_bill',
    y='tip',
    hue='time',      # encode meal time as colour
    size='size',     # encode party size as dot size
    alpha=0.7
)
plt.title('Tip vs Total Bill (coloured by Meal Time)')
plt.show()

Box Plots: Comparing Groups

A box plot shows the median, spread, and outliers across groups. If a feature's box looks very different per category, that feature is probably worth keeping.

import matplotlib.pyplot as plt
import seaborn as sns

df = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(12, 5))

sns.boxplot(data=df, x='day', y='total_bill', ax=axes[0])
axes[0].set_title('Bill by Day of Week')

sns.boxplot(data=df, x='smoker', y='tip', hue='sex', ax=axes[1])
axes[1].set_title('Tip by Smoking Status and Sex')

plt.tight_layout()
plt.show()

Correlation Heatmap: Finding Related Features

A correlation heatmap colors how strongly every pair of columns moves together. It reveals good predictors of your target — and redundant features to drop.

import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd

df = pd.read_csv('titanic.csv')

# Compute correlation matrix
corr = df[['Survived', 'Pclass', 'Age', 'SibSp', 'Parch', 'Fare']].corr()

# Plot heatmap
plt.figure(figsize=(8, 6))
sns.heatmap(
    corr,
    annot=True,     # show correlation values
    fmt='.2f',      # 2 decimal places
    cmap='RdYlGn',  # red-yellow-green colour scale
    vmin=-1, vmax=1
)
plt.title('Feature Correlation Matrix')
plt.show()

Pair Plots: Exploring All Feature Pairs

A pair plot shows scatter plots for every feature pair at once — a fast first look at a new dataset. Color by class to see which features separate the groups.

import seaborn as sns
import matplotlib.pyplot as plt

# Use the iris dataset (classic ML benchmark)
df = sns.load_dataset('iris')

# Pair plot coloured by species
sns.pairplot(
    df,
    hue='species',
    diag_kind='kde',    # KDE on diagonal
    plot_kws={'alpha': 0.6}
)
plt.suptitle('Iris Dataset Pair Plot', y=1.02)
plt.show()

Bar Charts and Count Plots

A count plot shows how often each category appears — the quickest way to check class balance before training a classifier. Add hue to compare two categories.

import seaborn as sns
import matplotlib.pyplot as plt

df = sns.load_dataset('titanic')

fig, axes = plt.subplots(1, 2, figsize=(12, 5))

# Class distribution (check balance)
sns.countplot(data=df, x='survived', ax=axes[0])
axes[0].set_title('Survival Count')

# Class by passenger class and sex
sns.countplot(data=df, x='class', hue='sex', ax=axes[1])
axes[1].set_title('Class Distribution by Sex')

plt.tight_layout()
plt.show()

Plotting Learning Curves

A learning curve plots training vs validation score as data grows. Both low means underfitting; a big gap means overfitting; both high and close means a good fit.

import matplotlib.pyplot as plt
import numpy as np
from sklearn.model_selection import learning_curve
from sklearn.tree import DecisionTreeClassifier
from sklearn.datasets import load_breast_cancer

X, y = load_breast_cancer(return_X_y=True)
train_sizes, train_scores, val_scores = learning_curve(
    DecisionTreeClassifier(max_depth=5), X, y, cv=5
)

plt.plot(train_sizes, train_scores.mean(axis=1), label='Training Score')
plt.plot(train_sizes, val_scores.mean(axis=1), label='Validation Score')
plt.xlabel('Training Set Size')
plt.ylabel('Accuracy')
plt.legend()
plt.title('Learning Curve')
plt.show()

Visualising Model Predictions

After training, plot the predictions. A confusion matrix heatmap shows where a classifier confuses labels — far more telling than a single accuracy number.

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix
from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split

X, y = load_iris(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)

model = DecisionTreeClassifier(max_depth=3)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

cm = confusion_matrix(y_test, y_pred)
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion Matrix')
plt.show()

Saving and Customising Plots

Good plots need polish: titles, axis labels, and readable fonts. Save them with savefig at dpi=150+ for reports, and pick a colorblind-friendly palette.

import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

# Create figure and axes explicitly
fig, ax = plt.subplots(figsize=(8, 5))

x = np.linspace(0, 10, 100)
ax.plot(x, np.sin(x), label='sin(x)', linewidth=2)
ax.plot(x, np.cos(x), label='cos(x)', linewidth=2, linestyle='--')

# Customise
ax.set_xlabel('x', fontsize=13)
ax.set_ylabel('y', fontsize=13)
ax.set_title('Sine and Cosine', fontsize=15, fontweight='bold')
ax.legend(fontsize=12)
ax.grid(True, alpha=0.3)

# Save
fig.savefig('plot.png', dpi=150, bbox_inches='tight')
plt.show()

Distribution Plots with Seaborn

A violin plot blends a box plot with a density curve, showing the full shape of a distribution. Seaborn's displot and kdeplot are great for smooth comparisons.

import seaborn as sns
import matplotlib.pyplot as plt

df = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(12, 5))

# Violin plot
sns.violinplot(data=df, x='day', y='total_bill', hue='sex',
               split=True, inner='quart', ax=axes[0])
axes[0].set_title('Bill Distribution by Day (Violin)')

# KDE distribution comparison
sns.kdeplot(data=df, x='tip', hue='time', fill=True, alpha=0.4, ax=axes[1])
axes[1].set_title('Tip Distribution by Meal Time (KDE)')

plt.tight_layout()
plt.show()

Quick Check

Test your understanding of Machine Learning with Python concepts from this lesson.

Lesson Recap

You learned to see your data: histograms and scatter plots reveal shape, heatmaps find predictors, and learning curves diagnose fit. Next: your first model! 🚀

常见问题解答

「使用 Matplotlib 和 Seaborn 可视化数据」课时是免费的吗？

是的 — 「使用 Matplotlib 和 Seaborn 可视化数据」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Machine Learning Academy 课程的其余内容，请升级到 CoddyKit PRO。 Machine Learning Academy 课程共包含 4 节课。

「使用 Matplotlib 和 Seaborn 可视化数据」这节课中我会学到什么？

您将绘制直方图、散点图和相关性热力图，在建模前探索数据分布与关系你通过在浏览器中直接运行的动手代码来练习 Machine Learning Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 Machine Learning Academy 需要有经验吗？

无需任何先前经验。CoddyKit 上的 Machine Learning Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 4 节课，共 4 节。

「使用 Matplotlib 和 Seaborn 可视化数据」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 Machine Learning Academy 课中编写并运行代码吗？

能。每节 Machine Learning Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 Machine Learning Academy