Search code examples
azureazure-machine-learning-service

Letter Recognition Error in Azure ML Studio

I'm having troubles with a Letter Recognition model I'm creating in Azure ML Studio.

I'm running a few algorithms - Decision Jungle, Neural Network, Decision Forest, Logistic Regression, One vs. All Multiclass, and then I append them using the Add rows method (Neural Network and Desicion Jungle/ Decision Forest and Logistic Regression), until I append them all.

However, appending Decision Forest and Logistic Regression I get the following error:

requestId = 9292bc066f51404eb5e0d0d219d3a072 errorComponent=Module. taskStatusCode=400. {"Exception":{"ErrorId":"NotInRangeValue","ErrorCode":"0008","ExceptionType":"ModuleException","Message":"Error 0008: Parameter \"Dataset2(number of columns)\" value should be in the range of [3, 3]."}}Error: Error 0008: Parameter "Dataset2(number of columns)" value should be in the range of [3, 3]. Process exited with error code -2

Any advice what should I do? Huge thanks in advance

Solution

  • This error occurs when there is a mismatch of number of columns of the two dataset you are appending.

    Looking at the error :

    The output of one model is returning rows with 3 columns and other one is having either more or less than 3 columns.

    Before this step "Add Rows" step -> Do quick Visualize

    enter image description here

    This will give a view of the dataset that you are planning to append.

    ![enter image description here

    Ensure for both, the columns numbers are same.