Evaluation
Explore how to evaluate machine learning models by calculating training and testing accuracy. Understand the importance of splitting data into training and test sets to identify overfitting and achieve generalized learning. This lesson guides you through coding a perceptron model evaluation to measure predictive performance and model reliability.
The ML process
For any complex problem that requires the computer to be able to identify patterns, there is an ML process to solve it.
This chapter demystifies each step of this process one by one. This lesson is about the last step—evaluation.
The problem that we wanted to solve was good movie classification. Instead of the traditional programmer’s approach, we used the machine learning approach. We trained the model on our data. We also know how to use the trained model once the training is over. This is the production mode. We have solved the pattern identification problem the machine learning way!
How good is our model?
However, have we really solved the problem? How do we know how good is our solution? If you notice, we do have the third column of labels in our dataset. These are the desired ...