#Pandas 数据可视化

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime, timedelta
import warnings
warnings.filterwarnings('ignore')

# 设置中文字体和负号显示
plt.rcParams['font.sans-serif'] = ['SimHei', 'Microsoft YaHei', 'DejaVu Sans']
plt.rcParams['axes.unicode_minus'] = False

# 设置图表样式
plt.style.use('seaborn-v0_8')
sns.set_palette("husl")

# 设置图表大小
plt.rcParams['figure.figsize'] = (12, 8)
plt.rcParams['font.size'] = 10

print("可视化环境配置完成")

# 设置随机种子
np.random.seed(42)

# 创建综合数据集
n_samples = 1000

# 销售数据
sales_data = {
    'date': pd.date_range('2022-01-01', periods=365, freq='D'),
    'product': np.random.choice(['产品A', '产品B', '产品C', '产品D', '产品E'], 365),
    'region': np.random.choice(['北区', '南区', '东区', '西区'], 365),
    'sales_amount': np.random.normal(1000, 300, 365),
    'quantity': np.random.randint(10, 100, 365),
    'customer_satisfaction': np.random.uniform(3.0, 5.0, 365)
}

# 确保销售额为正数
sales_data['sales_amount'] = np.maximum(sales_data['sales_amount'], 100)

sales_df = pd.DataFrame(sales_data)
sales_df['month'] = sales_df['date'].dt.month
sales_df['quarter'] = sales_df['date'].dt.quarter
sales_df['weekday'] = sales_df['date'].dt.day_name()

# 员工数据
employee_data = {
    'employee_id': range(1, n_samples + 1),
    'department': np.random.choice(['IT', 'HR', 'Finance', 'Marketing', 'Operations'], n_samples),
    'position': np.random.choice(['初级', '中级', '高级', '专家', '经理'], n_samples),
    'salary': np.random.normal(8000, 2000, n_samples),
    'age': np.random.randint(22, 60, n_samples),
    'experience_years': np.random.randint(0, 20, n_samples),
    'performance_score': np.random.uniform(60, 100, n_samples)
}

employee_data['salary'] = np.maximum(employee_data['salary'], 3000)
employee_df = pd.DataFrame(employee_data)

print("示例数据集创建完成")
print(f"销售数据: {sales_df.shape}")
print(f"员工数据: {employee_df.shape}")

print("\n=== 线图示例 ===")

# 时间序列线图
fig, axes = plt.subplots(2, 2, figsize=(15, 10))

# 1. 基本时间序列线图
daily_sales = sales_df.groupby('date')['sales_amount'].sum()
daily_sales.plot(ax=axes[0,0], color='blue', linewidth=2)
axes[0,0].set_title('每日销售额趋势')
axes[0,0].set_ylabel('销售额')
axes[0,0].grid(True, alpha=0.3)

# 2. 多条线图
product_daily_sales = sales_df.groupby(['date', 'product'])['sales_amount'].sum().unstack()
product_daily_sales.plot(ax=axes[0,1], marker='o', markersize=3)
axes[0,1].set_title('各产品每日销售额')
axes[0,1].set_ylabel('销售额')
axes[0,1].legend(title='产品', bbox_to_anchor=(1.05, 1), loc='upper left')
axes[0,1].grid(True, alpha=0.3)

# 3. 滚动平均线图
daily_sales_ma = daily_sales.rolling(window=7).mean()
axes[1,0].plot(daily_sales.index, daily_sales.values, alpha=0.3, label='原始数据')
axes[1,0].plot(daily_sales_ma.index, daily_sales_ma.values, color='red', linewidth=2, label='7日移动平均')
axes[1,0].set_title('销售额与移动平均')
axes[1,0].set_ylabel('销售额')
axes[1,0].legend()
axes[1,0].grid(True, alpha=0.3)

# 4. 面积图
region_daily_sales = sales_df.groupby(['date', 'region'])['sales_amount'].sum().unstack()
region_daily_sales.plot.area(ax=axes[1,1], alpha=0.7)
axes[1,1].set_title('各区域销售额堆叠面积图')
axes[1,1].set_ylabel('销售额')
axes[1,1].legend(title='区域', bbox_to_anchor=(1.05, 1), loc='upper left')

plt.tight_layout()
plt.show()

# 使用 Pandas 内置绘图方法
print("\n使用 Pandas 内置方法绘制线图:")
# 月度销售趋势
monthly_sales = sales_df.groupby('month')['sales_amount'].sum()
monthly_sales.plot(kind='line', marker='o', title='月度销售额趋势', figsize=(10, 6))
plt.ylabel('销售额')
plt.xlabel('月份')
plt.grid(True, alpha=0.3)
plt.show()

print("\n=== 柱状图示例 ===")

fig, axes = plt.subplots(2, 2, figsize=(15, 10))

# 1. 基本柱状图
product_sales = sales_df.groupby('product')['sales_amount'].sum().sort_values(ascending=False)
product_sales.plot(kind='bar', ax=axes[0,0], color='skyblue')
axes[0,0].set_title('各产品总销售额')
axes[0,0].set_ylabel('销售额')
axes[0,0].tick_params(axis='x', rotation=45)

# 2. 水平柱状图
region_sales = sales_df.groupby('region')['sales_amount'].sum().sort_values()
region_sales.plot(kind='barh', ax=axes[0,1], color='lightcoral')
axes[0,1].set_title('各区域总销售额')
axes[0,1].set_xlabel('销售额')

# 3. 分组柱状图
product_region_sales = sales_df.groupby(['product', 'region'])['sales_amount'].sum().unstack()
product_region_sales.plot(kind='bar', ax=axes[1,0])
axes[1,0].set_title('产品-区域销售额分组柱状图')
axes[1,0].set_ylabel('销售额')
axes[1,0].tick_params(axis='x', rotation=45)
axes[1,0].legend(title='区域', bbox_to_anchor=(1.05, 1), loc='upper left')

# 4. 堆叠柱状图
product_region_sales.plot(kind='bar', stacked=True, ax=axes[1,1])
axes[1,1].set_title('产品-区域销售额堆叠柱状图')
axes[1,1].set_ylabel('销售额')
axes[1,1].tick_params(axis='x', rotation=45)
axes[1,1].legend(title='区域', bbox_to_anchor=(1.05, 1), loc='upper left')

plt.tight_layout()
plt.show()

# 员工数据柱状图
print("\n员工数据柱状图:")
fig, axes = plt.subplots(1, 2, figsize=(15, 6))

# 部门人数分布
dept_count = employee_df['department'].value_counts()
dept_count.plot(kind='bar', ax=axes[0], color='lightgreen')
axes[0].set_title('各部门人数分布')
axes[0].set_ylabel('人数')
axes[0].tick_params(axis='x', rotation=45)

# 职位分布
position_count = employee_df['position'].value_counts()
position_count.plot(kind='pie', ax=axes[1], autopct='%1.1f%%')
axes[1].set_title('职位分布饼图')
axes[1].set_ylabel('')

plt.tight_layout()
plt.show()

print("\n=== 散点图示例 ===")

fig, axes = plt.subplots(2, 2, figsize=(15, 10))

# 1. 基本散点图
employee_df.plot.scatter(x='age', y='salary', ax=axes[0,0], alpha=0.6)
axes[0,0].set_title('年龄与薪资关系')
axes[0,0].set_xlabel('年龄')
axes[0,0].set_ylabel('薪资')

# 2. 带颜色分组的散点图
for dept in employee_df['department'].unique():
    dept_data = employee_df[employee_df['department'] == dept]
    axes[0,1].scatter(dept_data['experience_years'], dept_data['salary'], 
                     label=dept, alpha=0.6)
axes[0,1].set_title('工作经验与薪资关系（按部门分组）')
axes[0,1].set_xlabel('工作经验（年）')
axes[0,1].set_ylabel('薪资')
axes[0,1].legend()

# 3. 气泡图（散点大小表示第三个变量）
scatter = axes[1,0].scatter(employee_df['age'], employee_df['salary'], 
                           s=employee_df['performance_score']*2, 
                           c=employee_df['experience_years'], 
                           alpha=0.6, cmap='viridis')
axes[1,0].set_title('年龄-薪资-绩效-经验综合散点图')
axes[1,0].set_xlabel('年龄')
axes[1,0].set_ylabel('薪资')
colorbar = plt.colorbar(scatter, ax=axes[1,0])
colorbar.set_label('工作经验（年）')

# 4. 销售数据散点图
sales_df.plot.scatter(x='quantity', y='sales_amount', 
                     c='customer_satisfaction', 
                     colormap='RdYlBu', ax=axes[1,1], alpha=0.6)
axes[1,1].set_title('销量与销售额关系（颜色表示客户满意度）')
axes[1,1].set_xlabel('销量')
axes[1,1].set_ylabel('销售额')

plt.tight_layout()
plt.show()

# 相关性矩阵散点图
print("\n相关性分析散点图:")
numeric_cols = ['age', 'salary', 'experience_years', 'performance_score']
fig, ax = plt.subplots(figsize=(10, 8))
pd.plotting.scatter_matrix(employee_df[numeric_cols], ax=ax, alpha=0.6, diagonal='hist')
plt.suptitle('员工数据相关性矩阵散点图')
plt.tight_layout()
plt.show()

print("\n=== 直方图和密度图示例 ===")

fig, axes = plt.subplots(2, 2, figsize=(15, 10))

# 1. 基本直方图
employee_df['salary'].plot.hist(bins=30, ax=axes[0,0], alpha=0.7, color='skyblue')
axes[0,0].set_title('薪资分布直方图')
axes[0,0].set_xlabel('薪资')
axes[0,0].set_ylabel('频数')

# 2. 密度图
employee_df['salary'].plot.density(ax=axes[0,1], color='red', linewidth=2)
axes[0,1].set_title('薪资分布密度图')
axes[0,1].set_xlabel('薪资')
axes[0,1].set_ylabel('密度')

# 3. 分组直方图
for dept in employee_df['department'].unique():
    dept_salary = employee_df[employee_df['department'] == dept]['salary']
    axes[1,0].hist(dept_salary, alpha=0.5, label=dept, bins=20)
axes[1,0].set_title('各部门薪资分布对比')
axes[1,0].set_xlabel('薪资')
axes[1,0].set_ylabel('频数')
axes[1,0].legend()

# 4. 多变量密度图
for dept in employee_df['department'].unique():
    dept_salary = employee_df[employee_df['department'] == dept]['salary']
    dept_salary.plot.density(ax=axes[1,1], alpha=0.7, label=dept)
axes[1,1].set_title('各部门薪资密度分布对比')
axes[1,1].set_xlabel('薪资')
axes[1,1].set_ylabel('密度')
axes[1,1].legend()

plt.tight_layout()
plt.show()

# 使用 seaborn 绘制更美观的分布图
print("\n使用 Seaborn 绘制分布图:")
fig, axes = plt.subplots(1, 2, figsize=(15, 6))

# 直方图 + 密度图
sns.histplot(data=employee_df, x='salary', hue='department', 
            kde=True, ax=axes[0], alpha=0.6)
axes[0].set_title('各部门薪资分布（直方图+密度图）')

# 小提琴图
sns.violinplot(data=employee_df, x='department', y='salary', ax=axes[1])
axes[1].set_title('各部门薪资分布（小提琴图）')
axes[1].tick_params(axis='x', rotation=45)

plt.tight_layout()
plt.show()

print("\n=== 箱线图示例 ===")

fig, axes = plt.subplots(2, 2, figsize=(15, 10))

# 1. 基本箱线图
employee_df.boxplot(column='salary', ax=axes[0,0])
axes[0,0].set_title('薪资分布箱线图')
axes[0,0].set_ylabel('薪资')

# 2. 分组箱线图
employee_df.boxplot(column='salary', by='department', ax=axes[0,1])
axes[0,1].set_title('各部门薪资分布箱线图')
axes[0,1].set_xlabel('部门')
axes[0,1].set_ylabel('薪资')

# 3. 多变量箱线图
employee_df[['salary', 'age', 'performance_score']].boxplot(ax=axes[1,0])
axes[1,0].set_title('多变量箱线图')
axes[1,0].tick_params(axis='x', rotation=45)

# 4. 使用 seaborn 绘制更美观的箱线图
sns.boxplot(data=employee_df, x='position', y='salary', ax=axes[1,1])
axes[1,1].set_title('各职位薪资分布箱线图')
axes[1,1].tick_params(axis='x', rotation=45)

plt.tight_layout()
plt.show()

# 销售数据箱线图
print("\n销售数据箱线图:")
fig, axes = plt.subplots(1, 2, figsize=(15, 6))

# 各产品销售额箱线图
sns.boxplot(data=sales_df, x='product', y='sales_amount', ax=axes[0])
axes[0].set_title('各产品销售额分布')
axes[0].tick_params(axis='x', rotation=45)

# 各区域销售额箱线图
sns.boxplot(data=sales_df, x='region', y='sales_amount', ax=axes[1])
axes[1].set_title('各区域销售额分布')

plt.tight_layout()
plt.show()

print("\n=== 热力图示例 ===")

fig, axes = plt.subplots(2, 2, figsize=(15, 12))

# 1. 相关性热力图
corr_matrix = employee_df[['age', 'salary', 'experience_years', 'performance_score']].corr()
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', center=0, ax=axes[0,0])
axes[0,0].set_title('员工数据相关性热力图')

# 2. 透视表热力图
product_region_pivot = sales_df.pivot_table(
    values='sales_amount', 
    index='product', 
    columns='region', 
    aggfunc='mean'
)
sns.heatmap(product_region_pivot, annot=True, fmt='.0f', cmap='YlOrRd', ax=axes[0,1])
axes[0,1].set_title('产品-区域平均销售额热力图')

# 3. 时间序列热力图
sales_df['day'] = sales_df['date'].dt.day
month_day_sales = sales_df.pivot_table(
    values='sales_amount',
    index='month',
    columns='day',
    aggfunc='mean'
)
sns.heatmap(month_day_sales, cmap='viridis', ax=axes[1,0], cbar_kws={'label': '平均销售额'})
axes[1,0].set_title('月份-日期销售额热力图')
axes[1,0].set_xlabel('日期')
axes[1,0].set_ylabel('月份')

# 4. 部门-职位薪资热力图
dept_position_salary = employee_df.pivot_table(
    values='salary',
    index='department',
    columns='position',
    aggfunc='mean'
)
sns.heatmap(dept_position_salary, annot=True, fmt='.0f', cmap='plasma', ax=axes[1,1])
axes[1,1].set_title('部门-职位平均薪资热力图')

plt.tight_layout()
plt.show()

# 缺失值热力图
print("\n缺失值分析热力图:")
# 创建带缺失值的数据
df_with_missing = employee_df.copy()
np.random.seed(42)
for col in ['salary', 'age', 'performance_score']:
    missing_idx = np.random.choice(df_with_missing.index, size=50, replace=False)
    df_with_missing.loc[missing_idx, col] = np.nan

# 绘制缺失值热力图
plt.figure(figsize=(10, 6))
sns.heatmap(df_with_missing.isnull(), cbar=True, yticklabels=False, cmap='viridis')
plt.title('数据缺失值分布热力图')
plt.show()

print("\n=== 多子图布局示例 ===")

# 创建复杂的多子图布局
fig = plt.figure(figsize=(20, 15))

# 使用 GridSpec 创建自定义布局
from matplotlib.gridspec import GridSpec
gs = GridSpec(3, 4, figure=fig, hspace=0.3, wspace=0.3)

# 第一行：时间序列分析
ax1 = fig.add_subplot(gs[0, :2])
daily_sales = sales_df.groupby('date')['sales_amount'].sum()
daily_sales.plot(ax=ax1, color='blue', linewidth=2)
ax1.set_title('每日销售额趋势', fontsize=14)
ax1.grid(True, alpha=0.3)

ax2 = fig.add_subplot(gs[0, 2:])
monthly_sales = sales_df.groupby('month')['sales_amount'].sum()
monthly_sales.plot(kind='bar', ax=ax2, color='orange')
ax2.set_title('月度销售额分布', fontsize=14)
ax2.tick_params(axis='x', rotation=0)

# 第二行：产品和区域分析
ax3 = fig.add_subplot(gs[1, 0])
product_sales = sales_df.groupby('product')['sales_amount'].sum()
product_sales.plot(kind='pie', ax=ax3, autopct='%1.1f%%')
ax3.set_title('产品销售额占比', fontsize=14)
ax3.set_ylabel('')

ax4 = fig.add_subplot(gs[1, 1])
region_sales = sales_df.groupby('region')['sales_amount'].sum()
region_sales.plot(kind='bar', ax=ax4, color='lightcoral')
ax4.set_title('区域销售额对比', fontsize=14)
ax4.tick_params(axis='x', rotation=45)

ax5 = fig.add_subplot(gs[1, 2:])
sns.boxplot(data=sales_df, x='product', y='sales_amount', ax=ax5)
ax5.set_title('各产品销售额分布', fontsize=14)
ax5.tick_params(axis='x', rotation=45)

# 第三行：员工数据分析
ax6 = fig.add_subplot(gs[2, 0])
employee_df.plot.scatter(x='age', y='salary', alpha=0.6, ax=ax6)
ax6.set_title('年龄与薪资关系', fontsize=14)

ax7 = fig.add_subplot(gs[2, 1])
dept_count = employee_df['department'].value_counts()
dept_count.plot(kind='bar', ax=ax7, color='lightgreen')
ax7.set_title('部门人数分布', fontsize=14)
ax7.tick_params(axis='x', rotation=45)

ax8 = fig.add_subplot(gs[2, 2:])
corr_matrix = employee_df[['age', 'salary', 'experience_years', 'performance_score']].corr()
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', center=0, ax=ax8)
ax8.set_title('员工数据相关性', fontsize=14)

plt.suptitle('综合数据分析仪表板', fontsize=16, y=0.98)
plt.show()

print("\n=== 交互式可视化示例 ===")

try:
    import plotly.express as px
    import plotly.graph_objects as go
    from plotly.subplots import make_subplots
    
    print("创建交互式图表...")
    
    # 1. 交互式散点图
    fig_scatter = px.scatter(
        employee_df, 
        x='age', 
        y='salary',
        color='department',
        size='performance_score',
        hover_data=['experience_years'],
        title='员工年龄-薪资交互式散点图'
    )
    fig_scatter.show()
    
    # 2. 交互式时间序列图
    daily_sales_df = sales_df.groupby('date')['sales_amount'].sum().reset_index()
    fig_line = px.line(
        daily_sales_df,
        x='date',
        y='sales_amount',
        title='每日销售额交互式趋势图'
    )
    fig_line.show()
    
    # 3. 交互式柱状图
    product_sales_df = sales_df.groupby('product')['sales_amount'].sum().reset_index()
    fig_bar = px.bar(
        product_sales_df,
        x='product',
        y='sales_amount',
        title='产品销售额交互式柱状图'
    )
    fig_bar.show()
    
    # 4. 交互式热力图
    fig_heatmap = px.imshow(
        corr_matrix,
        text_auto=True,
        aspect="auto",
        title='相关性交互式热力图'
    )
    fig_heatmap.show()
    
except ImportError:
    print("Plotly 未安装，跳过交互式可视化演示")
    print("可以使用 'pip install plotly' 安装 Plotly")

# 创建动态更新的图表示例
def create_animated_chart():
    """创建动画图表"""
    try:
        import plotly.express as px
        
        # 创建月度数据
        monthly_data = []
        for month in range(1, 13):
            month_data = sales_df[sales_df['month'] == month]
            for product in month_data['product'].unique():
                product_data = month_data[month_data['product'] == product]
                monthly_data.append({
                    'month': month,
                    'product': product,
                    'total_sales': product_data['sales_amount'].sum(),
                    'avg_satisfaction': product_data['customer_satisfaction'].mean()
                })
        
        monthly_df = pd.DataFrame(monthly_data)
        
        # 创建动画散点图
        fig = px.scatter(
            monthly_df,
            x='total_sales',
            y='avg_satisfaction',
            animation_frame='month',
            color='product',
            size='total_sales',
            title='月度产品销售额与客户满意度动画图'
        )
        
        fig.show()
        
    except ImportError:
        print("需要安装 Plotly 才能创建动画图表")

# create_animated_chart()

print("\n=== 图表样式定制 ===")

# 创建专业的图表样式
def create_professional_chart():
    """创建专业样式的图表"""
    
    # 设置专业配色方案
    colors = ['#2E86AB', '#A23B72', '#F18F01', '#C73E1D', '#592E83']
    
    fig, axes = plt.subplots(2, 2, figsize=(16, 12))
    
    # 1. 专业柱状图
    product_sales = sales_df.groupby('product')['sales_amount'].sum().sort_values(ascending=False)
    bars = axes[0,0].bar(product_sales.index, product_sales.values, color=colors)
    axes[0,0].set_title('产品销售额排名', fontsize=16, fontweight='bold', pad=20)
    axes[0,0].set_ylabel('销售额 (万元)', fontsize=12)
    axes[0,0].tick_params(axis='x', rotation=45)
    
    # 添加数值标签
    for bar in bars:
        height = bar.get_height()
        axes[0,0].text(bar.get_x() + bar.get_width()/2., height,
                      f'{height:.0f}',
                      ha='center', va='bottom', fontsize=10)
    
    # 美化网格
    axes[0,0].grid(True, alpha=0.3, linestyle='--')
    axes[0,0].set_axisbelow(True)
    
    # 2. 专业线图
    monthly_sales = sales_df.groupby('month')['sales_amount'].sum()
    axes[0,1].plot(monthly_sales.index, monthly_sales.values, 
                  marker='o', linewidth=3, markersize=8, color=colors[0])
    axes[0,1].fill_between(monthly_sales.index, monthly_sales.values, alpha=0.3, color=colors[0])
    axes[0,1].set_title('月度销售趋势', fontsize=16, fontweight='bold', pad=20)
    axes[0,1].set_ylabel('销售额 (万元)', fontsize=12)
    axes[0,1].set_xlabel('月份', fontsize=12)
    axes[0,1].grid(True, alpha=0.3, linestyle='--')
    axes[0,1].set_axisbelow(True)
    
    # 3. 专业散点图
    for i, dept in enumerate(employee_df['department'].unique()):
        dept_data = employee_df[employee_df['department'] == dept]
        axes[1,0].scatter(dept_data['age'], dept_data['salary'], 
                         label=dept, alpha=0.7, s=60, color=colors[i % len(colors)])
    
    axes[1,0].set_title('年龄与薪资关系分析', fontsize=16, fontweight='bold', pad=20)
    axes[1,0].set_xlabel('年龄', fontsize=12)
    axes[1,0].set_ylabel('薪资 (元)', fontsize=12)
    axes[1,0].legend(frameon=True, fancybox=True, shadow=True)
    axes[1,0].grid(True, alpha=0.3, linestyle='--')
    axes[1,0].set_axisbelow(True)
    
    # 4. 专业饼图
    region_sales = sales_df.groupby('region')['sales_amount'].sum()
    wedges, texts, autotexts = axes[1,1].pie(region_sales.values, 
                                            labels=region_sales.index,
                                            autopct='%1.1f%%',
                                            colors=colors,
                                            explode=(0.05, 0.05, 0.05, 0.05),
                                            shadow=True,
                                            startangle=90)
    
    axes[1,1].set_title('区域销售额分布', fontsize=16, fontweight='bold', pad=20)
    
    # 美化饼图文字
    for autotext in autotexts:
        autotext.set_color('white')
        autotext.set_fontweight('bold')
    
    # 调整整体布局
    plt.tight_layout()
    plt.suptitle('销售数据分析报告', fontsize=20, fontweight='bold', y=0.98)
    plt.show()

create_professional_chart()

print("\n=== 主题和配色方案 ===")

# 定义多种配色方案
color_schemes = {
    'business': ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd'],
    'pastel': ['#AEC7E8', '#FFBB78', '#98DF8A', '#FF9896', '#C5B0D5'],
    'dark': ['#2E3440', '#3B4252', '#434C5E', '#4C566A', '#5E81AC'],
    'vibrant': ['#FF6B6B', '#4ECDC4', '#45B7D1', '#96CEB4', '#FFEAA7']
}

def apply_theme(theme_name='business'):
    """应用指定主题"""
    colors = color_schemes.get(theme_name, color_schemes['business'])
    
    fig, axes = plt.subplots(2, 2, figsize=(15, 10))
    
    # 应用主题色彩
    product_sales = sales_df.groupby('product')['sales_amount'].sum()
    product_sales.plot(kind='bar', ax=axes[0,0], color=colors)
    axes[0,0].set_title(f'{theme_name.title()} 主题 - 产品销售额')
    
    region_sales = sales_df.groupby('region')['sales_amount'].sum()
    region_sales.plot(kind='pie', ax=axes[0,1], colors=colors, autopct='%1.1f%%')
    axes[0,1].set_title(f'{theme_name.title()} 主题 - 区域分布')
    
    # 多线图
    for i, product in enumerate(sales_df['product'].unique()):
        product_data = sales_df[sales_df['product'] == product]
        monthly_data = product_data.groupby('month')['sales_amount'].sum()
        axes[1,0].plot(monthly_data.index, monthly_data.values, 
                      marker='o', label=product, color=colors[i % len(colors)])
    
    axes[1,0].set_title(f'{theme_name.title()} 主题 - 月度趋势')
    axes[1,0].legend()
    axes[1,0].grid(True, alpha=0.3)
    
    # 散点图
    for i, dept in enumerate(employee_df['department'].unique()):
        dept_data = employee_df[employee_df['department'] == dept]
        axes[1,1].scatter(dept_data['age'], dept_data['salary'], 
                         label=dept, alpha=0.7, color=colors[i % len(colors)])
    
    axes[1,1].set_title(f'{theme_name.title()} 主题 - 年龄薪资')
    axes[1,1].legend()
    axes[1,1].grid(True, alpha=0.3)
    
    plt.tight_layout()
    plt.show()

# 展示不同主题
for theme in ['business', 'pastel', 'vibrant']:
    print(f"\n应用 {theme} 主题:")
    apply_theme(theme)

print("\n=== 数据故事讲述 ===")

def create_data_story():
    """创建完整的数据故事"""
    
    fig = plt.figure(figsize=(20, 24))
    gs = GridSpec(6, 3, figure=fig, hspace=0.4, wspace=0.3)
    
    # 标题页
    title_ax = fig.add_subplot(gs[0, :])
    title_ax.text(0.5, 0.5, '2022年度销售业绩分析报告', 
                 ha='center', va='center', fontsize=24, fontweight='bold')
    title_ax.text(0.5, 0.3, '基于Pandas数据分析的深度洞察', 
                 ha='center', va='center', fontsize=16, style='italic')
    title_ax.axis('off')
    
    # 第一部分：总体概况
    ax1 = fig.add_subplot(gs[1, 0])
    total_sales = sales_df['sales_amount'].sum()
    total_orders = len(sales_df)
    avg_order_value = total_sales / total_orders
    
    metrics = ['总销售额', '订单数量', '平均订单价值']
    values = [total_sales/10000, total_orders, avg_order_value]
    colors = ['#FF6B6B', '#4ECDC4', '#45B7D1']
    
    bars = ax1.bar(metrics, values, color=colors)
    ax1.set_title('2022年度业绩概览', fontsize=14, fontweight='bold')
    ax1.set_ylabel('数值')
    
    # 添加数值标签
    for i, (bar, value) in enumerate(zip(bars, values)):
        if i == 0:
            label = f'{value:.1f}万'
        elif i == 1:
            label = f'{value:.0f}'
        else:
            label = f'{value:.0f}元'
        ax1.text(bar.get_x() + bar.get_width()/2., bar.get_height(),
                label, ha='center', va='bottom', fontweight='bold')
    
    # 第二部分：时间趋势
    ax2 = fig.add_subplot(gs[1, 1:])
    daily_sales = sales_df.groupby('date')['sales_amount'].sum()
    monthly_avg = daily_sales.resample('M').mean()
    
    ax2.plot(daily_sales.index, daily_sales.values, alpha=0.3, color='lightblue', label='每日销售额')
    ax2.plot(monthly_avg.index, monthly_avg.values, linewidth=3, color='darkblue', 
            marker='o', markersize=8, label='月度平均')
    ax2.set_title('销售额时间趋势分析', fontsize=14, fontweight='bold')
    ax2.set_ylabel('销售额')
    ax2.legend()
    ax2.grid(True, alpha=0.3)
    
    # 第三部分：产品分析
    ax3 = fig.add_subplot(gs[2, 0])
    product_sales = sales_df.groupby('product')['sales_amount'].sum().sort_values(ascending=True)
    product_sales.plot(kind='barh', ax=ax3, color='lightcoral')
    ax3.set_title('产品销售额排名', fontsize=14, fontweight='bold')
    ax3.set_xlabel('销售额')
    
    # 第四部分：区域分析
    ax4 = fig.add_subplot(gs[2, 1])
    region_sales = sales_df.groupby('region')['sales_amount'].sum()
    wedges, texts, autotexts = ax4.pie(region_sales.values, labels=region_sales.index,
                                      autopct='%1.1f%%', startangle=90,
                                      colors=['#FF9999', '#66B2FF', '#99FF99', '#FFCC99'])
    ax4.set_title('区域销售分布', fontsize=14, fontweight='bold')
    
    # 第五部分：客户满意度分析
    ax5 = fig.add_subplot(gs[2, 2])
    satisfaction_by_product = sales_df.groupby('product')['customer_satisfaction'].mean()
    satisfaction_by_product.plot(kind='bar', ax=ax5, color='lightgreen')
    ax5.set_title('产品客户满意度', fontsize=14, fontweight='bold')
    ax5.set_ylabel('满意度评分')
    ax5.tick_params(axis='x', rotation=45)
    
    # 第六部分：相关性分析
    ax6 = fig.add_subplot(gs[3, :])
    # 创建相关性数据
    correlation_data = sales_df[['sales_amount', 'quantity', 'customer_satisfaction']].corr()
    im = ax6.imshow(correlation_data, cmap='RdBu', aspect='auto')
    ax6.set_xticks(range(len(correlation_data.columns)))
    ax6.set_yticks(range(len(correlation_data.columns)))
    ax6.set_xticklabels(correlation_data.columns)
    ax6.set_yticklabels(correlation_data.columns)
    ax6.set_title('销售指标相关性分析', fontsize=14, fontweight='bold')
    
    # 添加相关系数标注
    for i in range(len(correlation_data.columns)):
        for j in range(len(correlation_data.columns)):
            ax6.text(j, i, f'{correlation_data.iloc[i, j]:.2f}',
                    ha='center', va='center', fontweight='bold')
    
    plt.colorbar(im, ax=ax6, shrink=0.8)
    
    # 第七部分：关键洞察
    insights_ax = fig.add_subplot(gs[4, :])
    insights_text = """
    关键洞察与建议：
    
    1. 销售趋势：全年销售呈现稳定增长态势，第四季度表现尤为突出
    2. 产品表现：产品A和产品B占据主要市场份额，建议加大投入
    3. 区域分布：各区域发展相对均衡，可考虑差异化策略
    4. 客户满意度：整体满意度良好，但仍有提升空间
    5. 相关性发现：销售额与数量正相关，客户满意度影响复购率
    
    建议行动：
    • 加强产品A和B的市场推广
    • 提升客户服务质量，提高满意度
    • 优化供应链，确保产品质量稳定
    • 建立客户反馈机制，持续改进
    """
    
    insights_ax.text(0.05, 0.95, insights_text, transform=insights_ax.transAxes,
                    fontsize=12, verticalalignment='top', fontfamily='monospace',
                    bbox=dict(boxstyle='round', facecolor='lightgray', alpha=0.8))
    insights_ax.axis('off')
    
    # 第八部分：数据来源和方法
    method_ax = fig.add_subplot(gs[5, :])
    method_text = """
    数据来源与分析方法：
    • 数据来源：2022年1月1日至12月31日销售系统数据
    • 样本量：365个数据点，涵盖5个产品线和4个销售区域
    • 分析工具：Python Pandas + Matplotlib + Seaborn
    • 统计方法：描述性统计、相关性分析、趋势分析
    • 可视化：多维度图表展示，支持交互式探索
    """
    
    method_ax.text(0.05, 0.5, method_text, transform=method_ax.transAxes,
                  fontsize=10, verticalalignment='center', fontfamily='monospace',
                  bbox=dict(boxstyle='round', facecolor='lightyellow', alpha=0.8))
    method_ax.axis('off')
    
    plt.suptitle('数据驱动的业务洞察报告', fontsize=20, fontweight='bold', y=0.98)
    plt.show()

create_data_story()

print("\n=== 图表导出示例 ===")

import os

# 创建输出目录
output_dir = 'charts_output'
if not os.path.exists(output_dir):
    os.makedirs(output_dir)

def export_charts():
    """导出各种格式的图表"""
    
    # 创建示例图表
    fig, ax = plt.subplots(figsize=(10, 6))
    
    monthly_sales = sales_df.groupby('month')['sales_amount'].sum()
    monthly_sales.plot(kind='bar', ax=ax, color='skyblue')
    ax.set_title('月度销售额分析', fontsize=16, fontweight='bold')
    ax.set_ylabel('销售额')
    ax.set_xlabel('月份')
    ax.grid(True, alpha=0.3)
    
    # 导出为不同格式
    formats = ['png', 'pdf', 'svg', 'jpg']
    
    for fmt in formats:
        filename = f'{output_dir}/monthly_sales.{fmt}'
        plt.savefig(filename, format=fmt, dpi=300, bbox_inches='tight')
        print(f"图表已保存为: {filename}")
    
    plt.show()
    
    # 批量导出多个图表
    chart_configs = [
        {
            'data': sales_df.groupby('product')['sales_amount'].sum(),
            'kind': 'bar',
            'title': '产品销售额对比',
            'filename': 'product_sales'
        },
        {
            'data': sales_df.groupby('region')['sales_amount'].sum(),
            'kind': 'pie',
            'title': '区域销售分布',
            'filename': 'region_distribution'
        }
    ]
    
    for config in chart_configs:
        fig, ax = plt.subplots(figsize=(10, 6))
        
        if config['kind'] == 'bar':
            config['data'].plot(kind='bar', ax=ax, color='lightcoral')
        elif config['kind'] == 'pie':
            config['data'].plot(kind='pie', ax=ax, autopct='%1.1f%%')
            ax.set_ylabel('')
        
        ax.set_title(config['title'], fontsize=16, fontweight='bold')
        
        # 保存为PNG格式
        filename = f"{output_dir}/{config['filename']}.png"
        plt.savefig(filename, dpi=300, bbox_inches='tight')
        print(f"图表已保存为: {filename}")
        
        plt.close()  # 关闭图表以释放内存

export_charts()

# 创建图表模板
def create_chart_template(data, chart_type, title, filename):
    """创建标准化的图表模板"""
    
    fig, ax = plt.subplots(figsize=(12, 8))
    
    # 应用统一样式
    plt.style.use('seaborn-v0_8')
    
    if chart_type == 'line':
        data.plot(kind='line', ax=ax, linewidth=3, marker='o', markersize=8)
    elif chart_type == 'bar':
        data.plot(kind='bar', ax=ax, color='#2E86AB')
    elif chart_type == 'pie':
        data.plot(kind='pie', ax=ax, autopct='%1.1f%%', startangle=90)
        ax.set_ylabel('')
    
    ax.set_title(title, fontsize=18, fontweight='bold', pad=20)
    ax.grid(True, alpha=0.3)
    
    # 添加水印
    fig.text(0.99, 0.01, 'Generated by Pandas Analytics', 
            ha='right', va='bottom', alpha=0.5, fontsize=8)
    
    # 保存图表
    plt.savefig(f'{output_dir}/{filename}.png', dpi=300, bbox_inches='tight')
    plt.show()
    
    return fig

print("\n使用图表模板:")
create_chart_template(
    sales_df.groupby('month')['sales_amount'].sum(),
    'line',
    '月度销售趋势分析',
    'monthly_trend_template'
)

print("\n=== 大数据可视化优化 ===")

# 创建大数据集
large_sales_data = pd.concat([sales_df] * 100, ignore_index=True)
print(f"大数据集大小: {large_sales_data.shape}")

def optimize_large_data_viz(df, sample_size=1000):
    """优化大数据可视化"""
    
    import time
    
    # 方法1: 数据采样
    start_time = time.time()
    sampled_df = df.sample(n=min(sample_size, len(df)))
    sampled_df.plot.scatter(x='quantity', y='sales_amount', alpha=0.6, figsize=(10, 6))
    plt.title(f'采样可视化 (样本量: {len(sampled_df)})')
    plt.show()
    sampling_time = time.time() - start_time
    
    # 方法2: 数据聚合
    start_time = time.time()
    aggregated_df = df.groupby(['product', 'region']).agg({
        'sales_amount': 'mean',
        'quantity': 'mean',
        'customer_satisfaction': 'mean'
    }).reset_index()
    
    aggregated_df.plot.scatter(x='quantity', y='sales_amount', 
                              s=aggregated_df['customer_satisfaction']*20,
                              c='customer_satisfaction', colormap='viridis',
                              figsize=(10, 6))
    plt.title(f'聚合可视化 (数据点: {len(aggregated_df)})')
    plt.show()
    aggregation_time = time.time() - start_time
    
    print(f"采样方法耗时: {sampling_time:.3f} 秒")
    print(f"聚合方法耗时: {aggregation_time:.3f} 秒")
    
    return sampled_df, aggregated_df

sampled_data, aggregated_data = optimize_large_data_viz(large_sales_data)

print("\n=== 内存优化技巧 ===")

def memory_efficient_plotting():
    """内存高效的绘图方法"""
    
    # 1. 使用生成器避免创建大量中间变量
    def plot_generator(df, group_col):
        for group_name, group_data in df.groupby(group_col):
            yield group_name, group_data
    
    # 2. 分批处理大数据
    chunk_size = 1000
    fig, ax = plt.subplots(figsize=(12, 8))
    
    colors = plt.cm.Set3(np.linspace(0, 1, len(sales_df['product'].unique())))
    
    for i, (product, product_data) in enumerate(plot_generator(sales_df, 'product')):
        # 只绘制前100个点以节省内存
        sample_data = product_data.head(100)
        ax.scatter(sample_data['quantity'], sample_data['sales_amount'], 
                  label=product, alpha=0.6, color=colors[i])
    
    ax.set_title('内存优化散点图')
    ax.set_xlabel('销量')
    ax.set_ylabel('销售额')
    ax.legend()
    plt.show()
    
    # 3. 及时清理图表对象
    plt.close('all')  # 关闭所有图表
    
    # 4. 使用上下文管理器
    with plt.style.context('seaborn-v0_8'):
        fig, ax = plt.subplots(figsize=(10, 6))
        sales_df.groupby('month')['sales_amount'].sum().plot(ax=ax)
        ax.set_title('使用上下文管理器的图表')
        plt.show()
    # 样式会自动恢复

memory_efficient_plotting()

print("\n=== 图表选择指南 ===")

def chart_selection_guide()

print("\n=== 可视化设计原则 ===")

def visualization_principles():
    """可视化设计原则示例"""
    
    principles = {
        "简洁性原则": "避免不必要的装饰，突出数据本身",
        "准确性原则": "确保图表准确反映数据，避免误导",
        "一致性原则": "保持颜色、字体、样式的一致性",
        "可读性原则": "确保标签、图例清晰可读",
        "目的性原则": "根据分析目的选择合适的图表类型"
    }
    
    print("可视化设计五大原则:")
    for principle, description in principles.items():
        print(f"• {principle}: {description}")
    
    # 好的设计示例
    fig, axes = plt.subplots(1, 2, figsize=(15, 6))
    
    # 好的设计
    monthly_sales = sales_df.groupby('month')['sales_amount'].sum()
    axes[0].plot(monthly_sales.index, monthly_sales.values, 
                linewidth=2, marker='o', markersize=6, color='#2E86AB')
    axes[0].set_title('月度销售趋势', fontsize=14, fontweight='bold', pad=15)
    axes[0].set_xlabel('月份', fontsize=12)
    axes[0].set_ylabel('销售额 (万元)', fontsize=12)
    axes[0].grid(True, alpha=0.3, linestyle='--')
    axes[0].set_axisbelow(True)
    
    # 添加数据标签
    for i, v in enumerate(monthly_sales.values):
        if i % 2 == 0:  # 只显示部分标签避免拥挤
            axes[0].annotate(f'{v:.0f}', (monthly_sales.index[i], v), 
                           textcoords="offset points", xytext=(0,10), ha='center')
    
    axes[0].text(0.02, 0.98, '✓ 好的设计', transform=axes[0].transAxes, 
                fontsize=12, fontweight='bold', color='green',
                verticalalignment='top', bbox=dict(boxstyle='round', facecolor='lightgreen', alpha=0.3))
    
    # 不好的设计示例
    axes[1].plot(monthly_sales.index, monthly_sales.values, 
                linewidth=5, marker='s', markersize=12, color='red')
    axes[1].set_title('MONTHLY SALES TREND!!!', fontsize=16, color='red')
    axes[1].set_facecolor('yellow')
    axes[1].grid(True, alpha=0.8, linewidth=2, color='blue')
    
    # 添加过多装饰
    for i, v in enumerate(monthly_sales.values):
        axes[1].annotate(f'{v:.2f}万元', (monthly_sales.index[i], v), 
                        textcoords="offset points", xytext=(0,15), ha='center',
                        fontsize=8, rotation=45)
    
    axes[1].text(0.02, 0.98, '✗ 不好的设计', transform=axes[1].transAxes, 
                fontsize=12, fontweight='bold', color='red',
                verticalalignment='top', bbox=dict(boxstyle='round', facecolor='lightcoral', alpha=0.3))
    
    plt.tight_layout()
    plt.show()

visualization_principles()