smiuser/hpos-data
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

drop duplicates in consolidated data

Browse Source
This commit is contained in:
Pritimay Sarkar
2023-11-28 14:06:40 +05:30
parent f79687acf1
commit 17ffc8a854
1 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
cloud-functions/python/functions/consolidation.py
View File

@@ -74,9 +74,12 @@ def consolidation(request):
df = df[(df['testTime'] > start_date) & (df['testTime'] <= end_date)]
df = df.sort_values(by=['testTime'], ascending=False)
# df = df.drop(['resultData', "reportUploadTime", "userImageURL", "testType", "birthYear", "testStatus", "reportPath", "createdBy", "csvPath", "result", "mobileId", "resultRatio", "localFlag", "led2", "led1"], axis=1)
df = df[["_id", "calculatedRatio", "deviceId", "deviceRatio", "kitSerial", "led1Average", "led1Buffer", "led1Sample", "led2Average", "led2Buffer", "led2Sample", "led3Average", "led3Buffer", "led3Sample", "led4Average", "led4Buffer", "led4Sample", "deviceSerialNumber", "name", "testTime", "classificationResult", "abs427", "abs555", "predictedDenovixRatio", "prdClassification"]]
df.rename(columns={'deviceSerialNumber': "loginId", "calculatedRatio": "calibratedRatio"}, inplace = True)
df = df.reindex(sorted(df.columns), axis=1)
df = df[["_id", "classificationResult", "prdClassification", "predictedDenovixRatio", "calculatedRatio", "deviceRatio", "kitSerial", "abs1", "led1Average", "led1Buffer", "led1Sample", "abs2", "led2Average", "led2Buffer", "led2Sample", "abs3", "led3Average", "led3Buffer", "led3Sample", "abs4", "led4Average", "led4Buffer", "led4Sample", "deviceId", "deviceSerialNumber", "name", "testTime"]]
df.rename(columns={'deviceSerialNumber': "loginId"}, inplace = True)
# df = df.reindex(sorted(df.columns), axis=1)
df = df.drop_duplicates()
df.to_excel(writer, sheet_name = 'data', index=False)
writer.close()