如何用pandas库读取CSV文件并按列分组进行数据统计？

本文共计295个文字，预计阅读时间需要2分钟。

当然可以，请您提供需要改写的原文，我将帮您进行修改。

代码：

# coding=gbk
# 从HostWrites.csv读取数据并分组统计

import pandas
import datetime

print "\r\n从 HostWrites.csv 读取数据并分组统计\r\n"

excel = pandas.read_csv("D:\Program Files\CrystalDiskInfo8_2_0\Smart\KXG6AZNV512G TOSHIBA39HS1002TMFQ\HostWrites.csv", header=None)

dates = []
kv = {}
# for row in excel.iterrows():
for row in excel.tail(100).values:
time = pandas.to_datetime(row[0])
value = row[1]
date = datetime.datetime.strftime(time, "%Y-%m-%d")

if date not in kv:
dates.append(date)
kv[date] = value

last = None
delta = 0
lst = []
for date in dates:
if last is not None:
delta = kv[date] - last
last = kv[date]

today = datetime.datetime.strftime(datetime.datetime.now(), "%Y-%m-%d")
yesterday = datetime.datetime.strftime(datetime.datetime.now() + datetime.timedelta(-1), "%Y-%m-%d")

if delta > 50 or date == today or date == yesterday:
lst.append((date, kv[date], delta))

print "日期".ljust(12, " "), "固态硬盘写入量".ljust(16, " "), "增量"

index = 0
for item in lst:
index += 1
if len(lst) - index < 20:
print str(item[0]).ljust(13, " "), str(item[1]).ljust(15, " "), item[2]

raw_input("输入任意字符退出：")

View Code

输出：