How to save API response to csv with Python

Question

I'm getting facedetection data from an API in this form:

{"id":1,"ageMin":0,"ageMax":100,"faceConfidence":66.72220611572266,"emotion":"ANGRY","emotionConfidence":50.0'
b'2540969848633,"eyeglasses":false,"eyeglassesConfidence":50.38102722167969,"eyesOpen":true,"eyesOpenConfidence":50.20328140258789'
b',"gender":"Male","genderConfidence":50.462989807128906,"smile":false,"smileConfidence":50.15522384643555,"sunglasses":false,"sun'
b'glassesConfidence":50.446510314941406}]'

I'd like to save this to a csv-file like this:

id  ageMin  ageMax  faceConfidence
1   0       100     66

... and so on. I tried to do it this way:

response = requests.get(url, headers=headers)
    with open('detections.csv', 'w') as f:
        writer = csv.writer(f)
        for item in response:
            writer.writerow(str(item))

That puts every char in its own cell. I've also tried to use item.id, but that gives an error: AttributeError: 'bytes' object has no attribute 'id'.

Could someone point me to the right direction?

Your response is JSON-formatted. Python doesn't automatically turn JSON strings to objects, hence why you're getting that AttributeError. Parse the string with the json library before attempting to access the object's properties. — esqew
– esqew, Commented Aug 3, 2020 at 19:33
To get json response, use response.json() than reading and converting to json — bigbounty
– bigbounty, Commented Aug 3, 2020 at 19:35

Serhii · Accepted Answer · 2020-08-03 19:35:20Z

3

Maybe an overkill for a small task, but you can do the following:

convert JSON response (do not forget to check exceptions, etc.) to python dictionary

dic = response.json()
Create a dataframe, for example using pandas:

df = pandas.DataFrame(dic)
Save to csv omitting index:

df.to_csv('detections.csv', index=False, sep="\t")

answered Aug 3, 2020 at 19:35

Serhii

3151 silver badge5 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Dharman · Accepted Answer · 2020-08-03 19:37:08Z

You can do this relatively easily with the pandas and json libraries.

import pandas as pd
import json

response = """{
    "id": 1,
    "ageMin": 0,
    "ageMax": 100,
    "faceConfidence": 66.72220611572266,
    "emotion": "ANGRY",
    "emotionConfidence": 50.0,
    "eyeglasses": false,
    "eyeglassesConfidence": 50.38102722167969,
    "eyesOpen": true,
    "eyesOpenConfidence": 50.20328140258789,
    "gender": "Male",
    "genderConfidence": 50.462989807128906,
    "smile": false,
    "smileConfidence": 50.15522384643555,
    "sunglasses": false,
    "glassesConfidence":50.446510314941406
}"""

file = json.loads(doc)

json = pd.DataFrame({"data": file})
json.to_csv("response.csv")

This is the response formatted to csv.

,data
ageMax,100
ageMin,0
emotion,ANGRY
emotionConfidence,50.0
eyeglasses,False
eyeglassesConfidence,50.38102722167969
eyesOpen,True
eyesOpenConfidence,50.20328140258789
faceConfidence,66.72220611572266
gender,Male
genderConfidence,50.462989807128906
glassesConfidence,50.446510314941406
id,1
smile,False
smileConfidence,50.15522384643555
sunglasses,False

Collectives™ on Stack Overflow

How to save API response to csv with Python

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related