1 统计出911电话类型数量
import pandas as pd
import numpy as np
df = pd.read_csv('./911.csv')
# print(df.head())
# print(df.info())
# print(df['title'])
# print(df['title'].str.split(':').tolist())
temp_list = df['title'].str.split(':').tolist() # 把元素按,分隔开,然后装入列表
type_list = list(set(i[0] for i in temp_list)) # 循环列表,去重得出所有的类型装进列表中
# print(type_list)
# 创建一个全为零的列表
pd_zeros = pd.DataFrame(np.zeros((df.shape[0],len(type_list))),columns=type_list)
# print(pd_zeros)
# 往里面填充1
for i in type_list:
pd_zeros[i][df['title'].str.contains(i)] = 1# contains(i) 包含i的时候,返回True
print(pd_zeros)
print(pd_zeros.sum(axis=0))
使用分组聚合方式:
import pandas as pd
import numpy as np
df = pd.read_csv('./911.csv')
# print(df.head())
# print(df.info())
# print(df['title'])
# print(df['title'].str.split(':').tolist())
temp_list = df['title'].str.split(':').tolist() # 把元素按,分隔开,然后装入列表
type_list = [i[0] for i in temp_list] # 循环列表,去重得出所有的类型装进列表中
df['type'] = pd.DataFrame(np.array(type_list).reshape((df.shape[0],1)))
print(df)
print(df.groupby(by='type').count()['title'])
生成一段时间范围:
常见频率缩写:
pandas重采样:
统计数据中不同月份电话次数和可视化展示:
import pandas as pd
import numpy as np
df = pd.read_csv('./911.csv')
df['timeStamp'] = pd.to_datetime(df['timeStamp']) # 转为时间序列
df.set_index('timeStamp',inplace=True)
# print(df)
df_count = df.resample('M').count()['title']
# print(df_count)
import matplotlib.pyplot as plt
plt.figure(figsize=(20,8),dpi=80)
_x = df_count.index
_y = df_count.values
_x = [i.strftime("%Y-%m-%d") for i in _x]# 时间格式化
plt.plot(range(len(_x)),_y)
plt.xticks(range(len(_x)),_x,rotation=45)
plt.show()
不同月份,不同类型:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
df = pd.read_csv('./911.csv')
df['timeStamp'] = pd.to_datetime(df['timeStamp']) # 转为时间序列
# 增加列
temp_list = df['title'].str.split(':').tolist()
type_list = [i[0] for i in temp_list]
# print(type_list)
df['type'] = type_list
# print(df)
df.set_index('timeStamp',inplace=True)
#
# # print(df)
plt.figure(figsize=(20,8),dpi=80)
for type_name,type_data in df.groupby(by='type'):
df_count = type_data.resample('M').count()['title']
#
# # print(df_count)
_x = df_count.index
_y = df_count.values
_x = [i.strftime("%Y-%m-%d") for i in _x]# 时间格式化
plt.plot(range(len(_x)),_y)
plt.xticks(range(len(_x)),_x,rotation=45)
plt.show()
结果:
Periodindex:
就是把分开的字符串通过period index组合成pandas时间类型
以pm2.5的数据为列:
import pandas as pd
import matplotlib.pyplot as plt
df = pd.read_csv('./BeijingPM20100101_20151231.csv')
# print(df.head())
# 设置好时间索引
period = pd.PeriodIndex(year=df['year'],month=df['month'],day=df['day'],hour=df['hour'],freq='H')
# print(period)
df['datetime'] = period
df.set_index('datetime',inplace=True)
df = df.resample('7D').mean()
# print(df)
df['PM_US Post'].dropna()
data = df['PM_US Post']
# 可视化
_x = data.index
_x = [i.strftime('%Y-%m-%d') for i in _x]
_y = data.values
plt.figure(figsize=(20,8),dpi=80)
plt.xticks(range(0,len(_x),10),_x[::10],rotation=45)
plt.plot(range(len(_x)),_y)
plt.show()