Get_ads_spend_HourOfDay.py

Get_ads_KPIs_HourOfDay.py

"""
Copyright (C) Lyla Yang - All Rights Reserved
Unauthorized copying of this file, via any medium is strictly prohibited
Proprietary and confidential
Written by Lyla Yang <https://www.linkedin.com/in/lyla-yang-aa850080/>, May 2019
"""

"""
Reference of field attributes
https://developers.google.com/adwords/api/docs/appendix/reports#field-attributes
"""

import pandas as pd
import numpy as np
import datetime
import io
import os
import glob
import sys
from googleads import adwords
from datetime import datetime, timedelta

# Define the report date range: last 28 days including today
start=datetime.today().date().isoformat().replace("-", "")
end=datetime.now() + timedelta(days= - 28)
end= end.date().isoformat().replace("-","")

# create a dictionary to storge three ad accounts information
Ad_acc={'Brand-1':'Adwords account ID 01', 'Brand-3':'Adwords account ID 02','Brand-3':'Adwords account ID 03'}

def run_Hour0fDay_kip_report(key, acc_id, start_date, end_date):

    '''
    This functioin is using report_downloader = adwords_client.GetReportDownloader(version='v201809') and 'CAMPAIGN_PERFORMANCE_REPORT'
    for more detail, please visit: 
    input is the adwords account ID in dictionary items, start date, and end date, date format is yyymmdd
    out put is the campaign level reports
    '''
    
    '''
   Input variables: 
   1. key (brand name): from Dictionary Ad_acc. Type: string
   2. acc_id (adwords account ID): from Dictionary Ad_acc. Type: string
   3. start_date: report start date. Type: yyymmdd
   4. end_date: report end date. Type: yyymmdd
   Output: csv file 
    '''
    
# Define output as a string
    output= io.StringIO()

  # Initialize client object.
    adwords_client = adwords.AdWordsClient.LoadFromStorage('.../googleads.yaml')

    adwords_client.SetClientCustomerId(acc_id)

    report_downloader = adwords_client.GetReportDownloader(version='v201809')

    report_query = (adwords.ReportQueryBuilder()
                      .Select('CampaignId', 'CampaignName','Date', 'DayOfWeek','HourOfDay','Clicks',
                              'Impressions', 'Cost','Conversions','ConversionValue')
                      .From('CAMPAIGN_PERFORMANCE_REPORT')
                      .Where('CampaignStatus').In('ENABLED')
                      .During(end_date+ ','+start_date) 
                      .Build())

    report_downloader.DownloadReportWithAwql(report_query, 'CSV', output, skip_report_header=True,
              skip_column_header=False, skip_report_summary=True,
              include_zero_impressions=False)

    output.seek(0)
    
    types= { 'CampaignId':pd.np.int64, 'Clicks': pd.np.float64, 'Impressions': pd.np.float64,
             'Cost': pd.np.float64,'Conversions': pd.np.float64,'ConversionValue': pd.np.float64 }

    df = pd.read_csv(output,low_memory=False, dtype= types, na_values=[' --'])
    # delete the first and last column
    df['Brand']=key
    # micro amount 1000000
    df['Cost']=df.Cost/1000000
    print(df.head())
    return df

# retrive KPIs for each adword account and generate csv file named by the brand name
for k, v in Ad_acc.items():
    df=run_campaign_performance_report(k, v,start, end)
    df.to_csv('.../'+k+'_adwords_HourOfDay_kpi.csv')

# Optional.You can also integrate all accounts data together. Comment lines below if you don't need them.
path='.../HourOfDay Spend'
all_csv=glob.glob(path+'/*.csv', recursive=True)
list_df=(pd.read_csv(f) for f in all_csv)
all_df=pd.concat(list_df)
all_df.to_csv('.../all_adwords_HourOfDay_kpi.csv')

i've read your article very nice job so much simpler than the documentation i've done everything like i m suppose to the only problem is i've got errors in the last step as u can see in the screen shots and i

please if anyone can help me i appreciate it so much my email is [email protected]