Machine Learning with Python for Everyone

If you have been following along, you know that I am trying to become a data scientist. That means that I am reading more books that are probably healthy for a person to binge read. I wanted to highlight some of the books that I have been reading and some comments. I am going to list them in the order that I read them and I will close with the order that I should have read them in.

Data Science from Scratch – Joel Grus

This really was an excellent book to start with. It game me a good overview of what the field looks like and how to use the tools – Python. Not knowing Python did give me some challenges, but I was able to work around this with Google. Like most of the other books that I read it starts with some foundation in math, which was good because I have been out of school for more than a decade. Overall I think this was a good place to start.

Deep Learning with Python – Francois Chollet

After reading about Neural Networks and Deep Learning, I was hooked. I google for a book and this one popped up. I had heard of TensorFlow before, but I didn’t know about Keras. Keras is a helper library which makes things very easy. This book was written by the creator of Keras. This one is awesome and in depth. I learned a lot from it, but there was some fundamental things that were still a bit confusing even with my superior google sleuthing abilities. I am still reading that last few chapters because I felt I needed to take a break. Honestly my brain was full, but I am a glutten for punishment, so I moved to the next book.

Mathematics for Machine Learning

This one was not for the faint of heart. I read through the first few chapters, but just like the last book it only made my brain hurt more. It is really a textbook in disguise, but it is really good. I plan on getting back to it soon because some of the things that are missing for me are the more advanced math.

Deep Learning from Scratch – Seth Weidman

This was the best book yet. Neural Networks are really simple in concept, but I was still having a hard time seeing that. Seth takes a great approach to teaching each concept. He breaks them down using the math, code and a diagram. This approach really worked for me. The biggest part that really drove things home was writing a the neural network from scratch. Learning to Use Keras was good, but without the deeper understanding it was all a little too much voodoo for me. I give this one 5 stars. After reading Joel Grus book, I should have read this one second.