Machine learning libraries like scikit-learn have made it easier for data scientists to apply complex algorithms like linear regression and logistic regression. These libraries abstract away much of the complexity and allow users to focus on the task of building models and analyzing data. However, even though these libraries provide a great deal of convenience, it is still essential for data scientists to have a solid understanding of the underlying concepts and logic behind these algorithms.
The biggest reason why we need to understand the logic behind it is that we couldn't use libraries with full confidence without understanding the logic. The algorithms might produce the numbers, but we might misinterpret them if we don't understand the algorithm. On top of that, implementing the code you don't know is extremely stressful.
Another reason is that you can choose the right algorithm based on the situation. Let's say you want to predict the housing price with its size. Simple linear regression is enough. However, if you want to know the housing price with multiple features, you need to use multiple linear regression.
Overall, understanding algorithms is important so that we can choose and implement them properly.

Machine learning libraries like scikit-learn have made it easier for data scientists to apply complex algorithms like linear regression and logistic regression. These libraries abstract away much of the complexity and allow users to focus on the task of building models and analyzing data. However, even though these libraries provide a great deal of convenience, it is still essential for data scientists to have a solid understanding of the underlying concepts and logic behind these algorithms.

The biggest reason why we need to understand the logic behind it is that we couldn't use libraries with full confidence without understanding the logic. The algorithms might produce the numbers, but we might misinterpret them if we don't understand the algorithm. On top of that, implementing the code you don't know is extremely stressful.

Another reason is that you can choose the right algorithm based on the situation. Let's say you want to predict the housing price with its size. Simple linear regression is enough. However, if you want to know the housing price with multiple features, you need to use multiple linear regression.

Overall, understanding algorithms is important so that we can choose and implement them properly.

Why we need to learn Machine Learning when there are many libraries