如何使用pandas对超大csv文件进行快速拆分详解

pip install pandas

import pandas as pd

# 读取csv文件
df = pd.read_csv("../super_big.csv")

# 获取文件总行数
row_num = len(df)

# 确定每个小文件要包含的数据量
step = 400

for start in range(0, row_num, step):
    stop = start + step
    filename = "./small_{}-{}.csv".fORMat(start, stop)
    d = df[start: stop]
    print("Saving file : " + filename + ", data size : " + str(len(d)))
    d.to_csv(fname, index=None)

# 输出如下
# Saving file : ./small_0-500.csv, data size : 500
# Saving file : ./small_500-1000.csv, data size : 500

import pandas
df = pandas.read_csv('./super_big.csv')
type(df)

<class 'pandas.core.frame.DataFrame'>

# 返回第一行
print(df.loc[0])
# 返回第二行
print(df.loc[1])

 d = df[start: stop]

data = {
    "name": ["peter", "rose", "joe"],
    "career": ["teacher", "engineer", "doctor"]
}

df = pd.DataFrame(data)
print(df["name"])

#0    peter
#1     rose
#2      joe
#Name: name, dtype: object

df = pd.read_csv("YOUT_CSV_FILE.csv")

df.to_csv(fname, index=None)

data = {
    "name": ["peter", "rose", "joe"],
    "career": ["teacher", "engineer", "doctor"]
}

df = pd.DataFrame(data)
df.to_csv("a.csv")

# 文件内容如下，注意每行的开头自动添加了行索引，从0开始递增
,name,career
0,peter,teacher
1,rose,engineer
2,joe,doctor