Hyperparameters tuning on Microsoft.ML using visual studio

Taekyoung Kim 1 Reputation point
2021-05-17T08:53:39.573+00:00

First of all, I'm really thankful you to look at my question and develop this kind of great software and packages for machine learning.

I'm using visual studio embedded in Unity3D to build classification models using Microsoft.ML, and I successfully built several models following some guidelines on Youtube (https://www.youtube.com/watch?v=VJTpbZrBAeU&t=2754s).
I want to one step forward to build more accurate classification models by tuning hyperparameters of models as python does. For example, an SVM model has several hyperparameters such as C and gamma, and python provides grid search to optimize the hyperparmeters.
However, I cannot find some way to tune and manipulate hyperparameters of classification models from Microsoft.ML package in Visual studio, in fact, I can build models without tuning hyperparameters and just using default values of them, I guess.

So, I wonder that can I tune and manipulate hyperparameters of classification models in Microsoft.ML package in Visual studio.

In building a model, I did like "var pipeline = context.Transforms.NormalizeMeanVariance("parameter1").Append.context ... .Append(context.BinaryClassification.Trainers.LdSvm());", and I'm guessing I can input some value in the bracket of 'LdSvm()' to manipulate hyperparameters.

Thank you in advance.

.NET Machine learning
.NET Machine learning
.NET: Microsoft Technologies based on the .NET software framework.Machine learning: A type of artificial intelligence focused on enabling computers to use observed data to evolve new behaviors that have not been explicitly programmed.
158 questions
0 comments No comments
{count} votes

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.