KwickClips Artificial Intelligence · 41 sec · free
Is a lower MSE better?
Short answer. Yes. Lower error is better.
What data should you test on?
Unseen data only.
MSE, RMSE and splits
| Errors 2, 2, 2, 2: MSE = 4 |
| RMSE = square root = 2 |
| Test on unseen data only |
Remember
| Lower error is better |
| Train, validate, test |
How do you score a model that predicts marks? Use MSE and RMSE. Square each error and average: MSE is four. Its square root, RMSE, is two. Always test on unseen data. Lower is better. Train, validate, then test.
This clip is from the full lesson: Model Evaluation Metrics: Confusion Matrix, Accuracy, Precision, Recall and F1 — 8 minutes, with the tables, the quick answers and the whole lesson in text.
Useful for: CBSE Class 10 Artificial Intelligence (417), CBSE Class 10 Artificial Intelligence (417), CBSE Class 10 Artificial Intelligence (417), CBSE Class 10 Artificial Intelligence (417)
More KwickClips from this lesson
Why is it called a confusion matrix?38 sec
What is the F1 formula?43 sec
Can a 99% accurate model be useless?42 secVoice-over is AI-generated; the script is written and checked by Kajal Ma'am. Confirm anything you plan around against your official board document. We never ask for a password or an OTP.