19- How to Detect OVERFITTING in Machine Learning? скачать в хорошем качестве

19- How to Detect OVERFITTING in Machine Learning?

In this video, we see how to detect OVERFITTING in Machine Learning, in particular, we see if the U-Net model has overfitted the training data: ------------------------------------------------------------------------- 00:00 What we did in the previous video and what happens today! 00:33 Creating object of our class, normalising the data and divide the data into train/val split 00:59 Coding the training cycle 06:31 Coding the validation cycle 09:12 Model training and validation starts! 09:42 Visualizing the training and validation losses in order to detect OVERFITTING 10:55 How to stop overfitting? ------------------------------------------------------------------------- ⚡But What Is Overfitting in Machine Learning? Overfitting is a serious issue in machine learning since it results in poor model generalization. Overfitting is a notorious problem, where a machine learning model is trained so well on the training data, that all it learns the EXACT values of the training data in the training set! As a result, the model ends up JUST memorizing the training set, which leads to very poor generalization to the unseen data, for example, the test data! ⚡Underfitting, however, is when the machine learning model has failed to learn the training data well-enough! This is the polar opposite of overfitting! How to compare Underfitting and Overfitting? However, in the battle between overfitting vs. underfitting, which one wins! The answer is none of them!!! ⚡Ideally, we are looking for is a good fit, which means that the model is trained JUST enough to perform well on the training set, but NOT over-trained to the point of completely memorizing the training data and failing to generalize to the unseen test data! As a result, it is CRUCIAL for you to understand how to detect OVERFITTING in your Machine Learning model! ⚡Join the MLDawn Discord Server:   / discord   🔴 GitHub: https://github.com/MLDawn/MLDawn-Proj... 🌎 Website: https://www.mldawn.com/ 🕊 Twitter:   / mldawn2018   🔗 Linked In:   / mehran-bazargani-14b352176   Keep up the good work and good luck! 🤞🍀