import pandas as pd

# 創(chuàng)建一個(gè)包含缺失值的DataFrame
data = {'A': [1, 2, None, 4], 'B': [5, None, 7, 8]}
df = pd.DataFrame(data)

# 判斷每個(gè)元素是否是缺失值
print("判斷每個(gè)元素是否是缺失值：")
print(df.isnull())

# 判斷每列是否存在缺失值
print("\n判斷每列是否存在缺失值：")
print(df.isnull().any())

# 刪除包含缺失值的行
df_dropped_rows = df.dropna()

# 刪除包含缺失值的列
df_dropped_columns = df.dropna(axis=1)

print("刪除包含缺失值的行：")
print(df_dropped_rows)

print("\n刪除包含缺失值的列：")
print(df_dropped_columns)

# 用指定值填充缺失值
df_filled_value = df.fillna(value=0)

# 使用前一個(gè)非缺失值填充缺失值（向前填充）
df_forward_filled = df.ffill()

print("用指定值填充缺失值：")
print(df_filled_value)

print("\n向前填充缺失值：")
print(df_forward_filled)

# 替換指定值
df_replace = df.replace({None: -1, 7: 77})

print("替換指定值：")
print(df_replace)

# 根據(jù)條件分割元素
df_split = df[df['A'] < 3]

print("根據(jù)條件分割元素：")
print(df_split)

# 創(chuàng)建一個(gè)包含字符串的DataFrame
data_str = {'Text': ['apple', 'banana', 'orange', 'grape']}
df_str = pd.DataFrame(data_str)

# 判斷是否包含特定字符串
contains_apple = df_str['Text'].str.contains('apple')

# 提取字符串中的數(shù)字
df_str['Number'] = df_str['Text'].str.extract('(\d+)')

print("判斷是否包含特定字符串：")
print(contains_apple)

print("\n提取字符串中的數(shù)字：")
print(df_str)

# 轉(zhuǎn)換列的數(shù)據(jù)類型
df['A'] = df['A'].astype(float)

print("轉(zhuǎn)換列的數(shù)據(jù)類型：")
print(df.dtypes)

# 去重
df_duplicates_removed = df.drop_duplicates()

print("去重后的DataFrame：")
print(df_duplicates_removed)

# 自定義清洗函數(shù)：將所有元素加倍
def double_elements(x):
    return x * 2

df_custom_clean = df.applymap(double_elements)

print("自定義清洗函數(shù)后的DataFrame：")
print(df_custom_clean)

# 創(chuàng)建包含日期的DataFrame
data_date = {'Date': ['2022-01-01', '2022-02-01', '2022-03-01']}
df_date = pd.DataFrame(data_date)

# 將字符串轉(zhuǎn)換為日期類型
df_date['Date'] = pd.to_datetime(df_date['Date'])

# 提取日期中的年、月、日
df_date['Year'] = df_date['Date'].dt.year
df_date['Month'] = df_date['Date'].dt.month
df_date['Day'] = df_date['Date'].dt.day

print("處理日期數(shù)據(jù)：")
print(df_date)

# 使用線性插值填充缺失值
df_interpolated = df.interpolate()

print("使用線性插值填充缺失值：")
print(df_interpolated)

# 創(chuàng)建一個(gè)包含透 視表數(shù)據(jù)的DataFrame
data_pivot = {'Category': ['A', 'B', 'A', 'B'],
              'Value': [10, 20, 30, 40]}
df_pivot = pd.DataFrame(data_pivot)

# 利用透 視表進(jìn)行數(shù)據(jù)匯總
pivot_table = pd.pivot_table(df_pivot, values='Value', index='Category', aggfunc='sum')

print("透 視表數(shù)據(jù)匯總：")
print(pivot_table)

# 創(chuàng)建兩個(gè)DataFrame用于合并
df1 = pd.DataFrame({'ID': [1, 2, 3], 'Name': ['Alice', 'Bob', 'Charlie']})
df2 = pd.DataFrame({'ID': [1, 2, 4], 'Age': [25, 30, 35]})

# 根據(jù)ID列合并兩個(gè)DataFrame
merged_df = pd.merge(df1, df2, on='ID', how='left')

print("合并兩個(gè)DataFrame：")
print(merged_df)

# 針對(duì)大數(shù)據(jù)集進(jìn)行隨機(jī)采樣
df_sampled = df.sample(frac=0.5, random_state=42)

print("隨機(jī)采樣后的DataFrame：")
print(df_sampled)

# 創(chuàng)建包含異常值的DataFrame
data_outliers = {'Value': [100, 200, 300, 1000]}
df_outliers = pd.DataFrame(data_outliers)

# 定義異常值上下限
lower_limit = df_outliers['Value'].mean() - 2 * df_outliers['Value'].std()
upper_limit = df_outliers['Value'].mean() + 2 * df_outliers['Value'].std()

# 根據(jù)異常值上下限進(jìn)行過濾
df_no_outliers = df_outliers[(df_outliers['Value'] > lower_limit) & (df_outliers['Value'] < upper_limit)]

print("處理異常值后的DataFrame：")
print(df_no_outliers)

# 創(chuàng)建包含分組數(shù)據(jù)的DataFrame
data_grouped = {'Category': ['A', 'B', 'A', 'B'],
                'Value': [10, 20, 30, 40]}
df_grouped = pd.DataFrame(data_grouped)

# 按Category分組并計(jì)算平均值
grouped_mean = df_grouped.groupby('Category')['Value'].mean()

print("分組統(tǒng)計(jì)的平均值：")
print(grouped_mean)

# 自定義缺失值處理函數(shù)：用均值填充缺失值
def fill_na_with_mean(column):
    mean_value = column.mean()
    return column.fillna(mean_value)

# 應(yīng)用自定義函數(shù)進(jìn)行缺失值填充
df_custom_fillna = df.apply(fill_na_with_mean)

print("自定義缺失值處理后的DataFrame：")
print(df_custom_fillna)

# 創(chuàng)建包含多層索引的DataFrame
data_multiindex = {'Value': [10, 20, 30, 40],
                   'Category': ['A', 'B', 'A', 'B']}
df_multiindex = pd.DataFrame(data_multiindex)
df_multiindex.set_index(['Category', df_multiindex.index], inplace=True)

print("含有多層索引的DataFrame：")
print(df_multiindex)

# 根據(jù)多層索引進(jìn)行篩選
result = df_multiindex.loc['A']

print("\n根據(jù)多層索引篩選后的DataFrame：")
print(result)

# 創(chuàng)建時(shí)間序列數(shù)據(jù)
date_rng = pd.date_range(start='2022-01-01', end='2022-01-10', freq='D')
df_time_series = pd.DataFrame(date_rng, columns=['date'])
df_time_series['value'] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

# 使用滑動(dòng)窗口計(jì)算均值
df_time_series['rolling_mean'] = df_time_series['value'].rolling(window=3).mean()

print("使用滑動(dòng)窗口計(jì)算均值：")
print(df_time_series)

# 創(chuàng)建包含堆疊數(shù)據(jù)的DataFrame
data_stacked = {'A': [1, 2, 3],
                'B': [4, 5, 6]}
df_stacked = pd.DataFrame(data_stacked)

# 將數(shù)據(jù)從寬格式變?yōu)殚L(zhǎng)格式
df_long = pd.melt(df_stacked, var_name='Category', value_name='Value')

print("將數(shù)據(jù)從寬格式變?yōu)殚L(zhǎng)格式：")
print(df_long)