Python数据分析行业核心库

pythonimport pandas as pd# 1. 读取CSV数据df = pd.read_csv("data.csv")# 2. 查看数据基本信息print(df.info()) # 数据类型、非空值数量print(df.describe()) # 数值型数据统计描述# 3. 处理缺失值df = df.fillna(df.mean(numeric_only=True)) # 数值列用均值填充# 4. 分组聚合grouped_df = df.groupby("category")["value"].agg(["sum", "mean"]) # 按类别分组计算求和与均值# 5. 保存结果grouped_df.to_excel("result.xlsx", index=False)

pythonimport numpy as np# 1. 创建多维数组arr = np.array([[1, 2, 3], [4, 5, 6]]) # 2x3数组print(arr.shape) # 输出(2, 3)# 2. 数组运算arr_sum = arr + 2 # 广播机制：每个元素加2arr_mul = arr @ arr.T # 矩阵乘法（2x3 @ 3x2 = 2x2）# 3. 数值计算print(np.mean(arr)) # 计算数组均值print(np.linalg.inv(arr_mul)) # 计算矩阵逆# 4. 随机数生成np.random.seed(42) # 固定随机种子rand_arr = np.random.randn(2, 3) # 生成符合正态分布的随机数组

pythonimport matplotlib.pyplot as pltimport numpy as np# 设置中文显示plt.rcParams['font.sans-serif'] = ['SimHei']plt.rcParams['axes.unicode_minus'] = False# 1. 创建画布与子图fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))# 2. 绘制线图x = np.linspace(0, 10, 100)y = np.sin(x)ax1.plot(x, y, color='red', label='sin(x)')ax1.set_xlabel('x轴')ax1.set_ylabel('y轴')ax1.set_title('正弦函数趋势图')ax1.legend()ax1.grid(True)# 3. 绘制柱状图categories = ['A', 'B', 'C', 'D']values = [20, 35, 30, 25]ax2.bar(categories, values, color='blue', alpha=0.7)ax2.set_xlabel('类别')ax2.set_ylabel('数值')ax2.set_title('各类别数值对比图')# 4. 导出图表plt.tight_layout()plt.savefig('visualization.png', dpi=300)

pythonimport seaborn as snsimport pandas as pdimport matplotlib.pyplot as plt# 设置中文显示plt.rcParams['font.sans-serif'] = ['SimHei']plt.rcParams['axes.unicode_minus'] = False# 加载示例数据（Pandas DataFrame）df = sns.load_dataset('iris')# 1. 绘制散点图矩阵（展示多变量相关性）sns.pairplot(df, hue='species', markers=['o', 's', 'D'])plt.title('鸢尾花数据散点图矩阵')# 2. 绘制热力图（展示数值相关性矩阵）corr = df.corr(numeric_only=True)sns.heatmap(corr, annot=True, cmap='coolwarm', fmt='.2f')plt.title('特征相关性热力图')# 3. 绘制箱线图（展示分类数据分布）sns.boxplot(x='species', y='petal_length', data=df)plt.title('不同品种鸢尾花花瓣长度分布')plt.show()

pythonimport plotly.express as pximport pandas as pd# 加载示例数据df = px.data.iris()# 1. 绘制交互式散点图fig1 = px.scatter(df, x='sepal_width', y='sepal_length', color='species', size='petal_length', hover_data=['petal_width'], title='鸢尾花数据交互式散点图')fig1.show() # 在浏览器或Notebook中显示交互式图表# 2. 绘制交互式柱状图fig2 = px.bar(df, x='species', y='petal_length', color='species', barmode='group', title='不同品种鸢尾花花瓣长度对比')fig2.update_layout(xaxis_title='品种', yaxis_title='花瓣长度')fig2.show()# 3. 绘制3D交互式散点图fig3 = px.scatter_3d(df, x='sepal_length', y='sepal_width', z='petal_length',color='species', symbol='species', title='鸢尾花数据3D散点图')fig3.show()# 4. 保存为HTML文件fig1.write_html('interactive_scatter.html')

pythonfrom sklearn.datasets import load_irisfrom sklearn.model_selection import train_test_splitfrom sklearn.ensemble import RandomForestClassifierfrom sklearn.metrics import accuracy_score, classification_report# 1. 加载数据并拆分训练集/测试集iris = load_iris()X = iris.data # 特征y = iris.target # 标签X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)# 2. 初始化模型并训练model = RandomForestClassifier(n_estimators=100, random_state=42)model.fit(X_train, y_train)# 3. 模型预测y_pred = model.predict(X_test)# 4. 模型评估print(f"模型准确率：{accuracy_score(y_test, y_pred):.2f}")print("分类报告：")print(classification_report(y_test, y_pred, target_names=iris.target_names))# 5. 查看特征重要性feature_importance = pd.DataFrame({'feature': iris.feature_names,'importance': model.feature_importances_}).sort_values('importance', ascending=False)print("特征重要性：")print(feature_importance)

pythonimport statsmodels.api as smimport pandas as pdimport numpy as np# 生成示例数据（线性回归数据）np.random.seed(42)X = np.random.randn(100, 2) # 2个特征X = sm.add_constant(X) # 添加常数项（截距）y = 2 * X[:, 0] + 3 * X[:, 1] + np.random.randn(100) # 真实模型：y=2x1+3x2+噪声# 1. 构建线性回归模型（OLS）model = sm.OLS(y, X)results = model.fit()# 2. 输出详细统计结果print(results.summary()) # 包含系数、p值、R²、F检验等指标# 3. 假设检验（示例：检验系数是否为0）print(f"特征1系数的p值：{results.pvalues[1]:.4f}")print(f"特征2系数的p值：{results.pvalues[2]:.4f}")# 4. 预测X_new = sm.add_constant(np.random.randn(5, 2)) # 新数据y_pred = results.predict(X_new)print("新数据预测结果：")print(y_pred)