0
$\begingroup$

If we divide the data into training data, validation data, and testing data, I remember the lesson from Andrew Ng saying we use the validation data for hyperparameter tuning purpose. (you can see this article: https://towardsdatascience.com/why-do-we-need-a-validation-set-in-addition-to-training-and-test-sets-5cf4a65550e0)

My question is why not using training data for hyperparameter tuning since we have more data within?

Improve this question
$\endgroup$
3
  • $\begingroup$ What do you mean by using the training data for hyperparameter tuning? $\endgroup$ Commented Apr 13, 2022 at 3:01
  • $\begingroup$ Meaning find the best parameters for the model using training data (do prediction, find the parameters that gives lowest errors) @Dave $\endgroup$ Commented Apr 13, 2022 at 3:02
  • $\begingroup$ Perhaps you could give an example of how this would go: what hyperparameter parameter would you aim to tune, and how would you tune it using only the training data? $\endgroup$ Commented Apr 13, 2022 at 3:07

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.