其他分享
首页 > 其他分享> > panda quantile, group by, summary row to list

panda quantile, group by, summary row to list

作者:互联网

import pandas as pd
import numpy as np

function to get summary list names

def deleteDuplicatelist(listA):
#return list(set(listA))
return sorted(set(listA), key = listA.index)

FCData=pd.read_csv(r’/Users/samwang/Downloads/BYD FC-Export-ID-115213015836970-2021-09-14T00_00_00-2021-09-27T23_59_59-X5476-DEVELOPMENT5-Versions-2.csv’)

print(FCData[‘Max Load’].head())

print(FCData[[‘Max Extension’,“Max Load.1”]].describe())

print(FCData.columns)

print(FCData.head())

print(FCData[[‘Site’,‘Product’]])

FCSdata= FCData[[‘Site’,‘Product’,‘SerialNumber’,‘StartTime’,‘test_type’,‘Max Load’]]

print(FCSdata.describe())

set startime as time format , then change to year-month-day format.

FCData[‘dateFormat’]=pd.to_datetime(FCData[“StartTime”],format="%Y/%m/%d")
FCData[‘day’]=FCData[‘dateFormat’].dt.strftime(’%Y-%m-%d’)

print(FCData)

LG and Sm by SN

product =[]
for a in FCData[‘SerialNumber’]:
if “13H9” in a:
product.append(“SM”)
elif “13HC” in a:
product.append(“LG”)
else:
“abnormalSn”
FCData[“product”]=product
list1= (deleteDuplicatelist(product))

print(list1)

quantileFrame=FCData[‘Max Load.1’].groupby(by=(FCData[“product”])).quantile(q=[0.05, 0.1, 0.63])
quantileFrame=FCData.groupby(by=(FCData[“product”]))[‘Max Load.1’].agg([‘mean’,“max”,“quantile”])

quantileFrame=FCData.groupby(by=[“product”,‘test_type’,“day”])[‘Max Load.1’].agg([‘mean’,“min”,“quantile”,np.percentile(q=95)])

quantileFrame=FCData.groupby(by=[“product”,‘test_type’,“day”])[‘Max Load.1’].agg([np.min,np.mean,‘count’,lambda x: x.quantile(0.05),lambda x: x.quantile(0.1)])

quantileFrame=FCData.groupby(by=[“product”,‘test_type’,“day”])[‘Max Load.1’].quantile(q=[0.05, 0.1, 0.63])

print(quantileFrame)

FCTable = FCData.pivot_table(index=[‘test_type’,“day”],values=‘Max Load.1’,aggfunc=“mean”)

print(FCTable)

‘Station ID’,‘Test Pass/Fail Status’,‘EndTime’‘fixture_id’‘Max extension’ ,‘Max Extension’, ‘Max Load’

print(FCData[FCData[‘test_type’].str.contains(‘CCP_Bare’)][“Max Load.1”].quantile(q=[0.05,0.1,0.63]))

print(FCData.groupby(by=“test_type”)[‘Max Load.1’].agg(“quantile(0.3)”))

list_name =(FCData[‘test_type’].values.tolist())
list_name=deleteDuplicatelist(list_name)
day_list = FCData[‘day’].values.tolist()
day_list=deleteDuplicatelist(day_list)
print(list_name)
print((day_list))

for day in day_list:
for name in list_name:

    # print(name, FCData[FCData['test_type'].str.contains(name)]["Max Load.1"].quantile(q=[0.05, 0.1, 0.63]).values,sep=" ")
    dateoutput=FCData[(FCData['test_type'].str.contains(name)) & (FCData['day'].str.contains(day))]["Max Load.1"]
    # print(name, day, dateoutput.min(), dateoutput.quantile(q=[0.05, 0.1, 0.63]).values,sep=" ")

标签:group,Max,quantile,list,FCData,print,day
来源: https://blog.csdn.net/qq_38844711/article/details/120613375