Supervised Learning

In supervised learning, the models are trained using well-labeled training data. It means, in training data, there is an output value(called a label) associated with every input value.

So, the data supervises or teaches the model on finding the relationship between input and output variables so that we can predict the label for an input variable in the real world on behalf of the relationship found during training.

Think about how we would have learned to identify an apple. I guess someone would have shown us an apple and told us that this fruit is called an apple. We would have also identified several characteristics on seeing it such as its shape is some-like spherical, its color is red, etc.

In the same way, we train our model to identify apples. We train it on several images of apples(telling the model it is an apple) and also on images of some other fruits(telling the model it is not an apple) so that our model can distinguish the apple from other fruits.

After the training is completed, the model would have found several characteristics of apples, so we can use our model to predict whether a fruit is an apple or not. Mathematically, these characteristics are called relationship.

enter image description here

Previous Index Next

End-to-End ML Project- Beginner friendly

Supervised Learning

XP

Loading comments...