python可视化-密度图

 1、加载数据

import pandas as pd
import numpy as np
from sklearn.datasets import load_iris
import warnings
 
# 禁用所有警告信息
warnings.filterwarnings('ignore')

# 加载数据
iris = load_iris()
iris

 iris.keys()

df = pd.DataFrame(iris.data, columns=iris.feature_names)
df['target'] = iris.target
df.head()

2、基于matplotlib的密度图

from scipy.stats import gaussian_kde


# 根据数据集构造密度函数
density = gaussian_kde(df['sepal width (cm)'])
density.covariance_factor = lambda : .25
density._compute_covariance()

# 构造向量作为x轴刻度
xs = np.linspace(1.5, 5, 200)

# 初始化
plt.figure(figsize=(8, 6))

plt.plot(xs, density(xs))
plt.show()

3、基于seaborn的密度图

import seaborn as sns
import matplotlib.pyplot as plt

sns.set(font='SimHei', font_scale=0.8, style="darkgrid") # 解决Seaborn中文显示问题

# 构造子图
fig, ax = plt.subplots(2,2,constrained_layout=True, figsize=(12, 8))


# 密度图
ax_sub = sns.kdeplot(df['sepal width (cm)'], ax=ax[0][0])
ax_sub.set_title('密度图')

# 水平密度图
ax_sub = sns.kdeplot(y=df['sepal width (cm)'], color="skyblue", ax=ax[0][1])
ax_sub.set_title('水平密度图')

# 增加阴影
ax_sub = sns.kdeplot(df['sepal width (cm)'], fill=True, ax=ax[1][0])
ax_sub.set_title('增加阴影')

sns.kdeplot(data=df, x='sepal width (cm)', fill=True, bw_method=0.1, ax=ax[1][1]).set_title('降低带宽')

plt.show()

4、一图绘制多个变量

ax_sunb = sns.kdeplot(data=df, x='sepal width (cm)', fill=True, color='r', label='sepal width (cm)')
ax_sub = sns.kdeplot(df['sepal length (cm)'], fill=True, color="b", label='sepal length (cm)')
plt.legend()
# 修改x标签
plt.xlabel('speal width/speal length')

plt.show()

# alpha：修改透明度
sns.kdeplot(data=df, x='sepal width (cm)', hue="target", fill=True, common_norm=False, alpha=0.4)

# 绘制网格（共用坐标轴）
g = sns.FacetGrid(data=df, col='target', hue='target', col_wrap=3)
# 绘制密度图
g = g.map(sns.kdeplot,"sepal width (cm)", cut=0, fill=True, common_norm=False, alpha=1, legend=False)
# 每个网格的标题
g = g.set_titles("{col_name}")

plt.show()

# 绘制堆积图
sns.kdeplot(data=df, x='sepal width (cm)', hue='target', common_norm=False, multiple="fill", alpha=1)

plt.show()