pandas实例

1 统计不同月份不同类型电话次数

import pandas as pd
import numpy as np
from matplotlib import pyplot as plt

df = pd.read_csv("./911.csv")

#获取分类情况
temp_list = df["title"].str.split(":").tolist()
cate_list = list(set([i[0] for i in temp_list ]))

#df里面添加1列，按照这1列进行分类
df["cate"] = pd.DataFrame(np.array(cate_list).reshape((df.shape[0],1)))
df.groupby(by="cate").count()["title"]

#构造全为0的数组
zeros_df = np.DataFrame((np.zeros((df.shape[0],len(cate_list))),columns=cate_list)

#赋值1
for cate in cate_list:
	zero_df[cate][df["title"].str.contains(cate)] = 1

#赋值2(慢，不建议)
for i in range(df.shape[0]):
	zeros_df.loc[i,temp_list[i][0]] = 1

sum_set = zeros_df.sum(axis=0)

2 pandas时间序列

pd.date_range(start=,end=,periods=,freq=)

import pandas as pd
#每隔1天
pd.data_range(start="20171230",end="20180131",freq="D")
#每隔10天
pd.data_range(start="20171230",end="20180131",freq="10D")
#生成10个
pd.data_range(start="20171230",periods,freq="D")
#月份
freq="M"

freq="H"

pandas重采样 resample
将时间序列从一个频率转化为另一个频率进行处理的过程。

import pandas as pd
import numpy as np
from matplotlib import pyplot as plt

df = pd.read_csv("./911.csv")

df["timeStamp"] = pd.to_datetime(df["timeStamp"])
df.set_index("timeStamp",inplace=True)

#不同月份的电话次数
count_by_month = df.resample("M").count()["title"]
print(count_by_month)

#画图
_x = count_by_month.index
_y = count_by_month.values


_x = [i.strftime(%Y%m%d) for i in _x]

plt.plot(range(len_x),_y)
plt.xticks(range(len(_x)),_x,rotation=45)

plt.show()

#不同月份不同类型的电话次数
import pandas as pd
import numpy as np
from matplotlib import pyplot as plt

def plot_img(df,label):
	count_by_month = df.resample("M"),count()["title"]
	_x = count_by_month.index
	_y = count_by_month.values


	_x = [i.strftime(%Y%m%d) for i in _x]

	plt.plot(range(len_x),_y,label=label)
	plt.xticks(range(len(_x)),_x,rotation=45)

#把时间字符串转为时间类型设置为索引
df = pd.read_csv("./911.csv")
df["timeStamp"] = pd.to_datetime(df["timeStamp"])
df.set_index("timeStamp",inplace=True)

#添加列，表示分类
#获取分类情况
temp_list = df["title"].str.split(":").tolist()
cate_list = list(set([i[0] for i in temp_list ]))

#df里面添加1列，按照这1列进行分类
df["cate"] = pd.DataFrame(np.array(cate_list).reshape((df.shape[0],1)))
df.groupby(by="cate").count()["title"]

#进行分组
for group_name,group_data in df.group(by="cate"):
	plot_img(group_data,group_name)

plt.legend(loc="best")
plt.show()

3 5个城市，PM2.5随时间变化

import pandas
from matplotlib import pyplot as plt

file_path = './PM2.5/BeijingPM20100101_20151231'
df = pd.read_csv(file_path)
#把分开的时间字符串通过periodIndex方法转化为pandas时间类型
period = pd.PeriodIndex(year=df["year"],month=df["month"],day=df["day"],hour=df["hour"],freq="H")

df["datetime"] = period

#把datetime设置为索引
df.set_index("datetime",inplace=True)

#进行降采样
df.resample("7D").mean()

#处理缺失数据,删除
data = df["PM_US Post"].dropna()

#画图
_x = data.index
_x = [i.strftime(%Y%m%d)for i in _x]
_y = values

plt.figure(figsize=(20,8),dpi=80)
plt.plot(range(len(_x)),_y)
plt.xticks(range(0,len(_x),20),list(_x)[::20],rotation=45)

plt.show()