From the course: Automated ML.NET Training, Metrics, and Accuracy

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

Inspect pipeline values

Inspect pipeline values - ML.NET Tutorial

From the course: Automated ML.NET Training, Metrics, and Accuracy

Start my 1-month free trial Buy for my team

Inspect pipeline values

“

- [Instructor] Part of creating an accurate model is ensuring the data is processed correctly from start to finish. ML.NET allows for inspection of the pipeline data to troubleshoot the model training process. It allows you to inspect intermediate data in all the steps, loading, processing and model training. Inspecting these values allows you to see the values as the machine learning progresses. Data is initially loaded using IDataView. To view the data, convert IDataView to IEnnumerable using the CreateEnnumerable method. This is because the IDataView is lazily evaluated. To optimize performance, set the reuseRowObject to true. Doing so will lazily load the same object with the data of the current row as it's being evaluated, as opposed to creating a new object for each row in the dataset. However, if you only need access to a portion of the data or specific indices, you can use the CreateEnnumerable and set the reuseRowObject parameter value to false. This will create a new object…

Contents