Applying ML Algorithms

I am almost done with week 3 of this course and satisfactorily understood the concepts. I had a general doubt about how these algorithms are applied to real-life problems. In the course, we were taught a rigid mathematical approach to create functions for fitting and predicting data. Likewise, we also saw glimpses of available ML libraries like SkLearn for the same purpose. My doubt is, do these libraries automatically account for these mathematical steps, or is there something else working behind the scenes. I mean, what might be the possible differences in the accuracy scores and results, if we manually code these functions and apply them to our data and on the other hand, directly fit train-test data in the libraries?
Apologies for such an out-of-context question :sweat_smile:

Hi!

Yes the libraries / frameworks like SkLearn and Tensorflow will be built on the mathematical principles learned in class and more.
If you are curious about Tensorflow offers you can check their documentation Module: tf  |  TensorFlow Core v2.9.1
and Tensroflow is open source too so you can check how it is implemented and contribute to their source code too! GitHub - tensorflow/tensorflow: An Open Source Machine Learning Framework for Everyone

Hope this helps!

Thanks for the answer! I’ll surely check out the links…