tick data 轉換成 k Bar

轉換tick資料成 k bar，做個記錄
有筆 tick csv 資料如下
TX00Ticks.csv
Name,nPtr,nDate,nTimehms,nTimemillismicros,nBid,nAsk,nClose,nQty
TX00,119,20190624,150000,17000,10621.0,10623.0,10622.0,36
TX00,120,20190624,150000,38000,10621.0,10622.0,10621.0,2
TX00,121,20190624,150000,167000,10620.0,10622.0,10622.0,1
TX00,122,20190624,150000,199000,10620.0,10621.0,10620.0,1
TX00,123,20190624,150000,205000,10620.0,10621.0,10620.0,1
TX00,124,20190624,150000,304000,10620.0,10621.0,10620.0,1
TX00,125,20190624,150000,321000,10620.0,10621.0,10621.0,2

pandas
要對一排一排的 row 做處理的話，apply 要設定 axis=1

ref: pandas apply 的使用
http://jonathansoma.com/lede/foundations/classes/pandas%20columns%20and%20functions/apply-a-function-to-every-row-in-a-pandas-dataframe/

#斬大寫得更簡單!! https://ithelp.ithome.com.tw/articles/10223693

---

import pandas as pd
from datetime import datetime

tick=pd.read_csv('TX00Ticks.csv')

#將三個column 的時間 轉成文字合併後，再用 datetime 轉成 datetime格式
def toDatetime(row):    
    t=str(row['nDate']) + ' ' + str(row['nTimehms']).zfill(6) + ' ' + str(row['nTimemillismicros'])
    return datetime.strptime(t, '%Y%m%d %H%M%S %f')

tick.index=tick.apply(toDatetime, axis=1)

tx1k=pd.DataFrame()
# 開 高 低 收 量
# T 是分鐘, 5分K就用 5T
tx1k['Open']=tick['nClose'].resample('T', label='right').first()
tx1k['High']=tick['nClose'].resample('T', label='right').max()
tx1k['Low']=tick['nClose'].resample('T', label='right').min()
tx1k['Close']=tick['nClose'].resample('T', label='right').last()
tx1k['Volume']=tick['nQty'].resample('T', label='right').sum()