Hello everyone, from now I see that these courses are great to grasp fundemental concepts, however, I think it is a bit short-sided to grasp how to code. There are not much coding exercises, What do you think?
Welcome to the community. I guess the answer to this question really depends on what you are looking for in this specialization.
Let’s say that the courses teaches how to code the ML algorithms from scratch but only explains them superficially. In that case, if tomorrow you have to say give an interview, teach someone the concept, answer some theoretical questions or even modify the algorithm to perhaps perform better, you might not be able to do that. Now, this approach is contrarian to what the courses follow right now.
On the other hand, if we consider the current approach, which is ensuring that the theoretical concepts are grasped by every learner, in that case, once you have learnt a programming language, you can easily transform the pseudocode to code yourself. And it’s not just that, the courses provide a ton of ungraded labs to help you understand how different things are implemented. Moreover, the graded labs make sure that you grasp the most important programming concepts as well.
Lastly, the thing to note here is that, you can find a ton of tutorials on the web explaining how to code a certain model, but the number of tutorials that can teach you the fundamental concepts with a good depth, are far fewer in number. I hope this helps.
Specialization focuses more on knowing the concepts and the basic configuration of each model of machine learning and why this was used for concepts and these functions and what is the relationship of mathematics and calculus in artificial intelligence and interpretation such as why log was used in logistic registration and how to enhance the model that you built and then how to implement this model From scratch, I also give exercises on some artificial intelligence libraries, but there is no course that will collect all the models and functions in each library and give exercises on them. These things are acquired only by applying what I learned through projects, competitions, and reading notebooks to others. all of these to give a better intuition on how to deal with new data and also what is the appropriate model for this data in addition to how to build your own model by combining two models with each other or improving something in a model to give you the best accuracy and so on