Request to Machine Learning service on Bluemix using Python

Question

I'm struggling in find a way to send data and have back the prediction of a SPSS model deployed on Bluemix Machine Learning service.

I make a lot of test using request library in Python or with curl command but I did not succeed.

I'm too new to Bluemix to understand the service documentation.

Any help,

Thanks

I am puzzled by your comment about "too new to Bluemix to understand the service documentation." Can you elaborate? If you have suggestions for the documentation, you can always submit feedback here: https://ibmcloud.ideas.aha.io/ — William 'Bill' Wentworth, Aug 25 '17 at 18:28
You are right Bill,what I meant is that I lack of experience in using Bluemix services and then some concepts in the documentation are to me unknown — Filippo Leoncini, Aug 25 '17 at 21:36
Fair enough. Just wanted to be sure that the documentation for it needed to be improved to increase clarity. If you see something, please be sure to submit feedback. — William 'Bill' Wentworth, Aug 28 '17 at 13:23

score 0 · Accepted Answer · edited Aug 26 '17 at 08:06

I managed to pass data in and receive the prediction with the code as follows:

import requests, urllib3, json

access_key= "INSERT_ACCESS_KEY_HERE"
username = "INSERT_USERNAME_HERE"
password = "INSERT_PASSWORD_HERE"
headers = urllib3.util.make_headers(basic_auth='{}:{}'.format(username, password))

payload_online= {  "tablename": "INSERT_TABLENAME_HERE",  "header": [INSERT_TABLE_HEADERS_HERE],"data": [[INSERT_DATA_TO_USE_FOR_THE_PREDICTION_HERE]]}

url= 'https://ibm-watson-ml.mybluemix.net/pm/v1/score/INSERT_CONTEXTID_HERE?accesskey=INSERT_THE_ACCESS_KEY'

header = {'Content-Type': 'application/json', 'Authorization': "INSERT_TOKEN_HERE"}


response_online = requests.post(url, json=payload_online, headers=header)

print(response_online.text)

Request to Machine Learning service on Bluemix using Python

1 Answers1