Model file is not found for Registration of model in training Pipeline.

Mike Ubezzi 2,776 Reputation points
2020-05-13T00:22:22.56+00:00

"We want the model to automatically register model every time there is a new model. we created the model in the process and write it out to a pipeline data set.To persist it then we upload and read it for registration.

We are using ./output to send the file to output. The issue is that it cannot find it in the file path . How can we validate its existence? "

[Note: As we migrate from MSDN, this question has been posted by an Azure Cloud Engineer as a frequently asked question] Source: MSDN

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,965 questions
0 comments
Accepted answer
  1. Ramr-msft 17,741 Reputation points
    2020-05-13T09:04:19.693+00:00

    Can you verify that the script that is actually writing the model file to the location you expect:

    with open(model_name, 'wb') as file:
       joblib.dump(value = model, filename = os.path.join('./outputs/', model_name))

    Inside in your train python script, you just need to do something like this:

    persist the model to the local machine

    tf.saved_model.save(model,'./outputs/model/')

    register the model with run object

    run.register_model(model_name,'./outputs/model/')

    Source: MSDN

    0 comments

0 additional answers

Sort by: Most helpful
Most helpful Newest Oldest

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.