Question

我正在读取具有3行3列的.csv。我将latitude和longitude发送到Web服务以获取24小时的天气信息。编辑：这是一个样本量，生产数据将有1000行。

我的输入数据的结构是：

field id| latitude| longitude
345       44.79721  -95.1756
789       44.7963   -95.1809
012       44.7954   -95.1925

我每行发送3个请求，以接收hourly temperature，hourly relative humidity和“每小时降水量”

我的输出在24小时内应该是这样的。数据表示来自startHour的值，这表示下面的第一行，尽管我正在00:00:00到01:00:00调用API。我想将开始时间和日期作为我的时间值保留在输出中。

time               | field id | Temp | Humidity | Precip
2019-05-06 00:00:00   345       65.5      7.8      0.0
2019-05-06 01:00:00   345       62.5      5.7      0.1
2019-05-06 02:00:00   345       59.2      3.1      0.0
2019-05-06 00:00:00   789       55.5      7.5      0.0
2019-05-06 01:00:00   789       59.5      5.2      0.0
2019-05-06 02:00:00   789       53.2      3.8      0.0
2019-05-06 00:00:00   012       58.5      7.5      0.0
2019-05-06 01:00:00   012       54.3      7.7      0.0
2019-05-06 02:00:00   012       53.2      10.8     0.2

我应该在代码中进行哪些更改，以写入上面示例中指定的日期和数据结构？请原谅格式或繁琐的代码。这是我的输出示例：

这是我的代码。

import pandas as pd
import requests
import json
import config
from datetime import datetime, timedelta
from pandas import DataFrame

df = pd.read_csv(config.DATA_PATH / config.input_csv,
                 usecols=['field id', 'lat', 'lon'],
                 dtype=str)

for _, id_, lat, lon in df.itertuples():
     rng = pd.date_range(pd.Timestamp(config.dayVal + ' ' + "'06:00:00'"),
                    periods=24, freq='H')


        dtSeries =  pd.Series(rng.format())

        ddf = dtSeries.to_frame(name='Date')


        ddf['time'] = pd.to_datetime(ddf['Date'])
        dateDF = ddf['dates'] = ddf['time'].dt.date

        timeDF = ddf['dates'] = ddf['time'].dt.time


        startDayVal= dateDF[::2]
        endDayVal = dateDF[1::2]


        startTimeVal= timeDF[::2]
        endTimeVal = timeDF[1::2]


        startDayValStr = (startDayVal.to_string(index=False)).replace(' ','').split('\n')
        endDayValStr =(endDayVal.to_string(index=False)).replace(' ','').split('\n')


        startTimeValStr = (startTimeVal.to_string(index=False)).replace(' ','').split('\n')
        endTimeValStr = (endTimeVal.to_string(index=False)).replace(' ','').split('\n')

        timeStr = (timeDF.to_string(index=False)).replace(' ','').split('\n')


        for _, id_, lat, lon in df.itertuples():
            for t in timeStr:
                for startDate, endDate, startTime, endTime, t in zip (startDayValStr, endDayValStr, startTimeValStr, endTimeValStr, timeStr):
                    time_param = '?start='+ startDate +'T'+startTime + 'Z' + '&end='+ endDate + 'T' + endTime + 'Z'

                    hrPrecip = 'https://insight.api.wdtinc.com/hourly-precipitation/' + str(lat)+'/' + str(lon) + time_param + '&unit=inches'
                    hrTemp = 'https://insight.api.wdtinc.com/hourly-temperature/' + str(lat)+'/' + str(lon) + time_param + '&unit=fahrenheit'
                    hrHumidity =  'https://insight.api.wdtinc.com/hourly-relative-humidity/' + str(lat)+'/' + str(lon) + time_param 


                    resp_pre_hr = se_.get(hrPrecip, auth=(config.key, config.password), timeout=30) 
                    resp_temp_hr = se_.get(hrTemp, auth=(config.key, config.password), timeout=60)  
                    resp_rh_hr = se_.get(hrHumidity, auth=(config.key, config.password), timeout=90)

                    preValHrly = str(json.loads(resp_pre_hr.content)['series'][0]['value'])
                    tempValHrly = str(json.loads(resp_temp_hr.content)['series'][0]['value'])
                    humidityHrly = str(json.loads(resp_rh_hr.content)['series'][0]['value'])

                    hrly_df = pd.DataFrame({'dateTime' :[t], 'field id': [id_], 'HourlyPrecipIn': [preValHrly],'HourlyRH' : [humidityHrly], 'HourlyTempF' : [tempValHrly]})
                    hrly_df = hrly_df[['dateTime','field id','HourlyPrecipIn','HourlyRH', 'HourlyTempF']]
                    hrly_df.head()
                    hrlydfs = hrLylst.append(hrly_df)
                    pd.concat(hrLylst).to_csv(config.output_csv_hourly, encoding='utf-8', index=False)

熊猫复制行输出

0 个答案: