pandas常用命令:

import numpy as np
import pandas as pd

my_list = list('abcdefghi')
my_arr = np.arange(26)
my_dict = dict(zip(my_arr, my_list))

series = pd.Series(my_dict)
print('--------------------------series data:')
print(series.head())
df = series.to_frame('value')              # series中的值在DataFrame中的列名
df.reset_index(inplace=True)               # 重新设置index,并显示series中的index
print('--------------------------df data:')
print(df)

series_one = pd.Series(range(8))
series_two = pd.Series(list('hwb'))
series_three = pd.Series(['3a', np.NaN, '3c'])
df_new = pd.DataFrame({'column1': series_one, 'column2': series_two, 'column3': series_three})
print('--------------------------df_new data:')
print(df_new)


# 按标签
print('--------------------- 1. index = 1: ')
print(df_new.loc[1])                          # 行号
print('--------------------- 2. index = all, 只显示column1,column2:')
print('方式1:', df_new.loc[:, ['column1', 'column2']])  # :所有, [x,y] 取值数组
print('方式2:', df_new.iloc[:, 0:2])                    # :所有,  x:y 范围x->y,包括x,不包含y
print("--------------------- 3. index = 0,2 , column = column1:")
print(df_new.loc[[0, 2], ['column1']])        # [x,y] 取值数组
print("--------------------- 4. index = 0,2 , column = all:")
print(df_new.loc[[0, 2], :])
print("--------------------- 5. index = [1:3), column = all :")
print(df_new.iloc[1:3, :])
print("--------------------- 6. index = 1, column = 1:")
print('方式1:', df_new.at[1, 'column2'])
print('方式2:', df_new.loc[1, 'column2'])
print('方式3:', df_new.iloc[1, 1])
# 说明: .loc[row_name, column_name] 列名,  .iloc[row_index, column_index] 索引

# column1 > 2
print('--------------------- 7. column1的值大于2:')
print(df_new[df_new.column1 > 2])