DAY 7 复习日

导入需要的库

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
data = pd.read_csv('heart.csv')
type(data)

查看相关特征

data.columns
data.shape
data.head()

进行描述性统计

data.describe
data.dtypes
data.info()

查看有无缺失值

data.isnull()
data.isnull().sum()
# 无缺失值

分别查找连续特征和离散特征

continuous_features = data.select_dtypes(include=['float64', 'int64']).columns.tolist()
discrete_features = data.select_dtypes(include=['object']).columns.tolist()

print("连续特征:"，continuous_features)
print("离散特征:"，discrete_features)

绘制单个连续变量可视化

#绘制单个连续变量的直方图
plt.figure(figsize=(16, 12))
for i, feature in enumerate(continuous_features, 1):
    plt.subplot(4, 4, i)  # 4行4列的子图
    #同理可使用sns.violinplot()绘制小提琴图
    # sns.violinplot(data[feature])
    # sns.boxplot(data[feature])  # 绘制箱线图
    # sns.kdeplot(data[feature], fill=True)  # 绘制核密度估计图
    # sns.distplot(data[feature], kde=True)  # 绘制分布
    sns.histplot(data[feature], kde=True, bins=30)
    plt.title(f'Histogram of {feature}', fontsize=14)
    plt.xlabel(feature)
    plt.ylabel('Count')
    plt.xticks(rotation=45)  # 旋转x轴标签以避免重叠
 
plt.tight_layout()  # 自动调整子图间距
plt.subplots_adjust(
    left=0.1,      # 左边缘
    bottom=0.1,    # 底部边缘
    right=0.9,     # 右边缘
    top=0.92,      # 顶部边缘
    wspace=0.3,    # 水平间距
    hspace=0.5     # 垂直间距
)
 
plt.suptitle('Histograms of Continuous Features', fontsize=20, y=0.98, weight='bold')  # 设置总标题
plt.savefig('histograms.png', dpi=300, bbox_inches='tight')  # 保存图像
plt.show()  # 显示图像

绘制特征和标签关系可视化

plt.figure(figsize=(8, 6))
sns.violinplot(x='target', y='cp', data=data)
plt.title('cp vs. target')
plt.xlabel('target')
plt.ylabel('cp')
plt.show()

plt.figure(figsize=(8, 6))
sns.violinplot(x='target', y='restecg', data=data)
plt.title('restecg vs. target')
plt.xlabel('target')
plt.ylabel('restecg')
plt.show()

@浙大疏锦行