TextClassification

Implemented with Microsoft Machine Learning Services

SQL Database Tables and Stored Procedures


Below are the data tables that you will find in either the TextClassification_R or TextClassification_Py databases after deployment.

Table Description
dbo.Label_Names Names associated with each of the 20 labels
dbo.Metrics Metrics for the trained model
dbo.Model Trained model
dbo.News_Test News items in the test set
dbo.News_To_Score News items to score
dbo.News_Train News items for training the model

The following stored procedures are used in this solutions in either the TextClassification_R or TextClassification_Py databases:

Stored Procedure Description
dbo.evaluateEvaluates the performance of the model
dbo.initial_Run_Once_R
dbo.initial_Run_Once_Py
Runs the training workflow natively in SQL for this solution
dbo.scoreCreates features on the fly for the testing set and makes predictions
dbo.train_modelCreates features on the fly for the training set and train the multiclass logistic regression model