Underfitting and overfitting - classification

دوره: یادگیری عمیق با TensorFlow / فصل: Overfitting / درس 2

Underfitting and overfitting - classification

توضیح مختصر

  • زمان مطالعه 0 دقیقه
  • سطح خیلی سخت

دانلود اپلیکیشن «زوم»

این درس را می‌توانید به بهترین شکل و با امکانات عالی در اپلیکیشن «زوم» بخوانید

دانلود اپلیکیشن «زوم»

فایل ویدیویی

متن انگلیسی درس

Time for overclassification example the two main types of supervised learning are regression and classification

and we’re trying to show you the most concepts in terms of both regression and classification will do

the same here.

Here’s our example there are two categories.

One of them is cats and the other one is dogs.

A good model explaining all the data looks like this a quadratic function with a few errors following

the logic we saw in our previous example.

What would be an under-funded model.

Well of course a linear model linear models are not very smart.

Often a simple when your model under fits.

If the data is not transformed around 60 percent of the observations would be classified correctly with

an fitted model now and overfitting model would classify the observations perfectly right.

It has correctly identified all the cat and dog photos in the data set.

But once we give it a different data set following the same quadratic function logic it will perform


All right let’s conclude lesson with the following remark.

What does a well-trained model then.

Well it is somewhere between an underpinning and an overfitting model.

This fine balance is often called the bias variance tradeoff O.

And if you are a Meem fan there’s this Facebook page.

Machine learning Nemes for convolutional teens that I follow some time ago they posted a photo of a

bed that beautifully exemplifies overfitting.

It fits some people perfectly but on average missed the point of being a bed Mame’s or an interesting

way to check knowledge.

If you find it funny you most probably understand the concept.

As you can see we are back to our interesting conceptual lessons.

So stick around.

Thanks for watching.

مشارکت کنندگان در این صفحه

تا کنون فردی در بازسازی این صفحه مشارکت نداشته است.

🖊 شما نیز می‌توانید برای مشارکت در ترجمه‌ی این صفحه یا اصلاح متن انگلیسی، به این لینک مراجعه بفرمایید.