Bias-Variance Trade-off | Overfitting and Underfitting | Easily explained!! | Machine Learning
In this very important tutorial, we're going to talk about the very important concept of Bias-Variance Trade-off and how does it relate to the undisputed party-poopers of our machine learning models --- Overfitting and Underfitting.
These topics are so important that they should be grasped before learning how to fit a machine learning algorithm to a dataset and celebrate when the accuracy shows 98%!
In the tutorial, we'll be going through all the nitty-gritties of the trade-offs we need to make to avoid underfitting or overfitting of our precious models, how to spot an underfitting or overfitting model, how to rectify that scenario and much more with examples.
I've uploaded all the relevant code and datasets used here (and all other tutorials for that matter) on my github page which is accessible here:
Link:
github.com/rachittoshniwal/ma...
If you like my content, please don not forget to upvote this video and subscribe to my channel.
If you have any qualms regarding any of the content here, please feel free to comment below and I'll be happy to assist you in whatever capacity possible.
Thank you!
Пікірлер: 5
I love your videos - they are very systematic and you keep your statements conceptual and clear! I've really learned a lot from them. In this video though, at 14:25, you say a model will over-fit as it is trained on more examples. I think that is not true, or at least debatable. The way I think about it, when a model of a given complexity (n model-parameters) is exposed to more training data, it becomes less affected by outliers in the training data and so, less over-fitted. In fact, a standard approach for reducing variance is to actually train on more data. I think more data = more averaging out of outliers, and not more model complexity, which is a property of model architecture and is unaffected by the amount of training data.
amazing tutorial !!
@rachittoshniwal
3 жыл бұрын
Thank you so much! :)
Excellent. Keep up the good work
@rachittoshniwal
3 жыл бұрын
Thank you Hector!