接种月龄计算

 brk <- read_csv("/mnt/c/Users/xuefeng/Desktop/brk.csv",locale = locale(encoding = 'GB18030')) %>% 

  clean_names() %>% 

  mutate(jz_sj=ymd_hms(jz_sj),csrq=ymd(csrq),jzyl = interval(csrq, jz_sj) %/% months(1),shi=str_sub(gldw_bm,1,4))

brk %>% 

  filter(csrq <= ymd('2023-12-30') & jz_zc == 1) %>% 

  group_by(shi) %>% 

  summarise(count = n(),

            jzyl_3 = sum(jzyl == 3,na.rm = T),jzyl_3/count*100) %>% 

  writexl::write_xlsx("/mnt/c/Users/xuefeng/Desktop/brk2024_3.xlsx")

import pandas as pd

import numpy as np

import janitor

from pandas.tseries.offsets import MonthEnd

from datetime import datetime

brk = pd.read_csv("/mnt/c/Users/xuefeng/Desktop/brk.csv", encoding='GB18030')

brk = (

    brk

    .rename(columns=lambda x: x.lower().replace(' ', '_'))

    .assign(jz_sj=lambda df: pd.to_datetime(df['jz_sj'], format='%Y-%m-%d %H:%M:%S',errors='coerce'),

            csrq=lambda df: pd.to_datetime(df['csrq'], format='%Y-%m-%d',errors='coerce'))

    .assign(jzyl=lambda df: (df['jz_sj'] - df['csrq']) / pd.Timedelta(days=30.436875))

    .assign(jzyl=lambda df: df['jzyl'].apply(np.floor))

    .assign(shi=lambda df: df['gldw_bm'].astype(str).str.slice(0, 4))

)

(

    brk

    .query("csrq<='2023-12-31' & jz_zc==1")

    .groupby('shi')

    .agg(count=('shi', 'count'),

         jzyl_3=('jzyl', lambda x: sum(x == 3)))

    .reset_index()

    .assign(prep=lambda df:df['jzyl_3']/df['count']*100)

)