Heroku Python Flask 应用程序中的请求超时

Question

I have trained a DL model and trying to expose it on the internet via REST API by hosting it in Heroku server.

In local I have tried all the Heroku configuration, tested the API and then finally deployed it successfully on Heroku. While accessing that endpoint the request is getting timed out and in heroku log I can see the following output:

2019-09-22T11:11:10.543181+00:00 app[web.1]: [2019-09-22 11:11:10 +0000] [48] [DEBUG] POST /analyse
2019-09-22T11:11:10.556943+00:00 app[web.1]: prediction started
2019-09-22T11:11:40.537864+00:00 heroku[router]: at=error code=H12 desc="Request timeout" method=POST path="/analyse" host=obscure-peak-46855.herokuapp.com request_id=f8f450ca-fd07-491d-aba2-433da710a3e1 fwd="103.242.239.84" dyno=web.1 connect=0ms service=30001ms status=503 bytes=0 protocol=https
2019-09-22T11:11:41.382624+00:00 app[web.1]: [2019-09-22 11:11:41 +0000] [4] [CRITICAL] WORKER TIMEOUT (pid:48)
2019-09-22T11:11:42.402385+00:00 app[web.1]: [2019-09-22 11:11:42 +0000] [49] [INFO] Booting worker with pid: 49
2019-09-22T11:12:39.861556+00:00 app[web.1]: [2019-09-22 11:12:39 +0000] [49] [DEBUG] Closing connection.

the representing Python code:

statusDict = {}
print('prediction started')
predictionResult = 1 if model.predict(inputData)[0] > 0.5 else 0
print('prediction complete')
statusDict['input comment'] = inputSentence
statusDict['status'] = 'prediction complete'
statusDict['prediction'] = predictionResult
response.update(statusDict)

Did anyone faced similar request timed out while exposing your ML model on Heroku? I can't even see the prediction complete print statement also in Heroku log.

Answer 1

This is happening because your model is taking too much time to load. Can you tell us more about the slug size and model. This error comes when the overall flow of your model takes more than 30 sec to render the page. can you share the Procfile and app.py as well to see where it is taking the time. Try to remove unecessary libraries if any in the file.