Azure machine learning cannot evaluate model

Andrea 1

I create a trained model in Azure machine learning, when I use it to predict a new set of data, the evaluate model cannot show the result, it's all empty. How can I solve the problem?

romungi-MSFT 45,036 Reputation points Microsoft Employee

2022-04-18T08:00:32.783+00:00

@Andrea I believe you have trained your model using the Azure ML designer. Usually evaluate model is used to test the accuracy of the trained model with inputs from a dataset containing scores generated from a model or models where you can evaluate the results of more than one dataset. The visualization of output of this module is based on certain configurations that have been setup for the module. Is your configuration based on the guidance on how to setup evaluation of the model?
Also, In the above case are you trying to score your model by deploying it to an endpoint since you want to predict with new data?
romungi-MSFT 45,036 Reputation points Microsoft Employee

2022-04-26T09:47:38.113+00:00

@Andrea Did you get a chance to check the configuration of your modules and try the scoring endpoint?
zahra alsaleh 0 Reputation points

2024-09-18T19:01:01.9966667+00:00

the evaluation model failed. i make true in appending output , the arrow's to the predicted and actual data are correct . in the score model the actual and preicted are recognized

Share via

Azure machine learning cannot evaluate model

Your answer