python numpy、pandas之奇技淫巧

df['a'] = df['a'].map(lambda x: x*2)

ls = []  # 存放修改后的数据
df.loc[:,('股票代码')] = ls

df['table'] = ''

df['table'].fillna(0, inplace=True)

df.dropna(axis="index", how="any", inplace=True)  # 有为空的值，则丢弃
df.reset_index(drop=True, inplace=True)  # 重置索引

random = df.sample(n=2000, replace=False)  # 不放回抽取

df.loc[i, 'a']  # 访问第i行的'a'元素
df.loc[i, 'a':'g']  # 访问第i行的'a'到'g'元素
df.loc[i, 'a':'g'] = new_info  # 修改指定行
df.loc[index:index, ('table_1', 'table_2', 'table_3')] = [info_1, info_2, info_3]  # 修改同一行的多列数据

import pandas as pd
from sqlalchemy import create_engine

engine = create_engine('mysql+pymysql://root:你的密码@localhost:3306/你的数据库名称')
con = engine.connect()  # df是已有的Dataframe类型数据
df.to_sql('example', con=engine, index=False, if_exists='replace')

df = pd.read_csv(r'D:\Export_Output.csv')

for index, row in df.iterrows():
    print(row['info_1'], row['info_2'])  # 遍历指定行

df_lilst = df.values.tolist()

dict_data={}
ls = []  # 总列表
ls_name = []  # 表头

with open(path_txt,'r')as df:
    for line in df:
        if line.count('\n') == len(line):  # 换行符跳过
            continue
        for kv in [line.strip().split('\t')]:
            # print(kv)
            ls.append(kv)

for i in ls[0]:
    ls_name.append(i)

frame = DataFrame(ls,columns=ls_name)  # 转为dataframe
frame.to_csv('file_out0.csv',encoding='gbk',index=False,header=False)

for root, dirs, files in os.walk(path):
    for name in files:
        all_path = os.path.join(path, name)　　# 文件总路径
        w = open(all_path, 'w')
        w.writelines([item for item in lines[:-1]])  # 删除最后一行
        w.close()

for root, dirs, files in os.walk(path):
    for name in files:
        all_path = os.path.join(path, name)　　# 文件总路径
        readFile = open(all_path)
        lines = readFile.readlines()
		lines = open(all_path).readlines()
		open(all_path, 'w').writelines(lines[1:])  # 删除第一行

count += 1
print("\r当前进度: {:.2f}%".format(count * 100 / length), end="")  # 进度条显示