How can I have the API return a 500 status code on error?

[andb0t]

Hi everybody!

I am following the official tutorials and thus my run function looks like this:

@input_schema('data', StandardPythonParameterType(input_sample))
@output_schema(StandardPythonParameterType(output_sample))
def run(data):
    try:
        result = model.predict(data)
        return [result]
    except Exception as e:
        error = str(e)
        return error

So everything runs fine if the there is no error. But in case of error (e.g. the user inputing the data in the wrong format) the API returns a 200 status code as well accompanied by the string error message as output. This behavior is clear from reading the code. My question is now: how can I make the API return a 500 on error and output the error message?

I tried something like this, but then the 500 was simply included in the output string and it still returned a 200:

        return error, 500

This happily gives me the string ["some error message", 500] with a 200 status code :)

Any idea how I can achieve this?
Thanks in advance!

@andb0t Did you figure out how to do this?

[andb0t]

No, unfortunately I haven't found anything... It still still happily returns a 200 with the primitive error message as return string. We have chosen to live with this on the meantime. Makes life and debugging harder...

I have not been able to test it yet, but might test it later.
You could possibly look into using this:
from azureml.contrib.services.aml_response import AMLResponse

example:

from azureml.contrib.services.aml_request import AMLRequest, rawhttp
from azureml.contrib.services.aml_response import AMLResponse
from PIL import Image
import json


def init():
    print("This is init()")
    

@rawhttp
def run(request):
    print("This is run()")
    
    if request.method == 'GET':
        # For this example, just return the URL for GETs.
        respBody = str.encode(request.full_path)
        return AMLResponse(respBody, 200)
    elif request.method == 'POST':
        file_bytes = request.files["image"]
        image = Image.open(file_bytes).convert('RGB')
        # For a real-world solution, you would load the data from reqBody
        # and send it to the model. Then return the response.

        # For demonstration purposes, this example just returns the size of the image as the response..
        return AMLResponse(json.dumps(image.size), 200)
    else:
        return AMLResponse("bad request", 500)

Found at: https://learn.microsoft.com/en-us/azure/machine-learning/v1/how-to-deploy-advanced-entry-script