search

MicrosoftDocs / azure-docs

Open source documentation of Microsoft Azure
https://docs.microsoft.com/azure
Creative Commons Attribution 4.0 International
10.25k stars 21.42k forks source link

Issue with the experiment #21687

Closed TrishaChandra123 closed 5 years ago

TrishaChandra123 commented 5 years ago

I followed these steps with a different dataset. I keep getting an error when I deploy the predictive experiment and run the test. Please find the links to the training as well as predictive experiment here: https://gallery.cortanaintelligence.com/Experiment/TM-Training https://gallery.cortanaintelligence.com/Experiment/TM-Predictive-Exp I am not sure why it's expecting the n-grams as input features. The error says: "ScoredFeaturesMustMatchTrainingFeatures: table: The data set being scored must contain all features used during training, missing feature(s): 'Preprocessed text_column.[away]', 'Preprocessed text_column.[stop]', 'Preprocessed text_column.[study]', 'Preprocessed text_column.[award]', 'Preprocessed text_column.[head]', 'Preprocessed text_column.[hehe]', 'Preprocessed text_column.[hope_you]', 'Preprocessed text_column.[write]', 'Preprocessed text_column.[totally]', 'Preprocessed text_column.[confuse]', ......................."

Document Details

⚠ Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.

femsulu commented 5 years ago

@TrishaChandra123 Thanks for the Comment. We are actively investigating and will get back to you soon.

YutongTie-MSFT commented 5 years ago

@TrishaChandra123 Hi, it seems that the input features you are using is not the same the model trained. Please check that. I have checked the document, there is nothing wrong about it. I would recommend you to post your issue in our MSDN forum for questions about your project. Our community will take care of your question there. Thank you. https://social.msdn.microsoft.com/Forums/en-US/home?forum=MachineLearning

We are closing this post since it is not a document issue.