2017-06-13 7 views
0

ich den folgenden Code aus der Community Notebook mit Predict Outdoor-Ausrüstung Kauf mit IBM Watson Machine Learning:502 Bad Gateway - Aufruf Scoring api Endpunkt

... 
<code omitted for brevity> 
... 

import urllib3, requests, json 
​ 
headers = urllib3.util.make_headers(basic_auth='{}:{}'.format(username, password)) 
url = '{}/v2/identity/token'.format(service_path) 
response = requests.get(url, headers=headers) 
mltoken = json.loads(response.text).get('token') 

endpoint_online = service_path + "/v2/online/deployments/" 
header_online = {'Content-Type': 'application/json', 'Authorization': mltoken} 
payload_online = {"artifactVersionHref": saved_model.meta.prop("modelVersionHref"), "name": "Product Line Prediction"} 
​ 
response_online = requests.post(endpoint_online, json=payload_online, headers=header_online) 
​ 
print response_online 
print response_online.text 

scoring_href = json.loads(response_online.text).get('entity').get('scoringHref') 
print scoring_href 

Die Antwort

<Response [201]> 
{"metadata":{"guid":"4148","href":"https://ibm-watson-ml.mybluemix.net/v2/online/deployments/4148","createdAt":"2017-06-13T07:54:16.062Z","modifiedAt":"2017-06-13T07:54:16.062Z"},"entity":{"scoringHref":"https://ibm-watson-ml.mybluemix.net/32768/v2/scoring/4148"}} 
https://ibm-watson-ml.mybluemix.net/32768/v2/scoring/4148 

Nächster Versuch, ein Tor zu erzielen:

payload_scoring = {"record":["M", 23, "Single", "Student"]} 
response_scoring = requests.put(scoring_href, json=payload_scoring, headers=header_online) 
​ 
print response_scoring.text 

Die Antwort:

<html> 
<head><title>502 Bad Gateway</title></head> 
<body bgcolor="white"> 
<center><h1>502 Bad Gateway</h1></center> 
<hr><center>nginx/1.10.1</center> 
</body> 
</html> 

Antwort

0

ich erneut versucht den Anruf ein paar Minuten später, und der Anruf erfolgreich war:

{ 
"result":{ 
     "PROFESSION_IX":6.0, 
     "GENDER_IX":0.0, 
     "MARITAL_STATUS_IX":1.0, 
     "GENDER":"M", 
     "features":{ 
     "values":[ 
      0.0, 
      23.0, 
      1.0, 
      6.0 
     ] 
     }, 
     "predictedLabel":"Personal Accessories", 
     "prediction":1.0, 
     ... 
} 
Verwandte Themen