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

Unlock the full course today

Join today to access over 24,700 courses taught by industry experts.

Hyperparameter tuning

Hyperparameter tuning - ML.NET Tutorial

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

Start my 1-month free trial Buy for my team

Hyperparameter tuning

“

- [Instructor] Hyperparameter tuning is another effective method for improving model accuracy. It's the ability to set learning parameters for training a model and making it more accurate. Hyperparameters are top-level parameters that control the learning process and the model parameters that result from it. These hyperparameter values are set by the developer before training and remain the same when the training ends. These values include things such as the train-test ratio, algorithm optimization, and the number of iterations. So these parameters are fixed during and after the training, meaning they don't change after the training is completed. Parameters on the other hand, are variables that the model learns from during the training. These are values such as the weights, biases, and coefficients of a linear regression. Hyperparameters are fixed during training, but parameters are updated by the learning algorithm itself. Hyperparameter tuning can be done manually by trial and…

Contents