Loy GitHub Issue

**Category example**<br><br> • area-System.Net<br> • area-System.Diagnostics<br> • area-System.Memory<br> • area-System.Data<br> <br><br> **Working Steps**<br> <br> • Prepare your data<br> • Transform the data<br> • Train the model<br> • Evaluate the model<br> • Predict with the trained model<br> <br> **Dataset**<br> <br> Github issues test tab separated file (issues_test.tsv)<br> <br> • ID - issue's ID<br> • Area - issue's label (named this way to avoid confusion with the Label concept in ML.NET)<br> • Title - issue's title<br> • Description - issue's description<br> <br> ![Dataset example][1] <br><br> Metrics for multiclass classification Evaluation<br><br> • Micro Accuracy - Every sample-class pair cntributes equally to the accuracy metric. You want Micro Accuracy to be as close to 1 as possible.<br><br> • Macro Accuracy: Every class contributes equally to the accuracy metric. Minority classes are given equal weight as the larger classes. You want Macro Accuracy to be as close to 1 as possible.<br><br> • Log-loss: You want Log-loss to be as close to zero as possible.<br><br> • Log-loss reduction - Ranges from [-inf, 100], where 100 is perfect predictions and 0 indicates mean predictions. You want Log-loss reduction to be as close to zero as possible.<br><br> <br> ![enter image description here][2] <br> ![enter image description here][3] [1]: https://raw.githubusercontent.com/laploy/ML.NET/master/GitHub-Issue/github-issue-azureML-dataset.JPG [2]: https://raw.githubusercontent.com/laploy/ML.NET/master/GitHub-Issue/github-issue-azureML-model.JPG [3]: https://raw.githubusercontent.com/laploy/ML.NET/master/GitHub-Issue/github-issue-azureML-evall.JPG