Tutorial given during the OceanHackWeek 2020
In April 2018 I started Udacity’s Nanodegree in Machine Learning Engineer. The classes are not cheap and many questions asked me the same thing: does it worth it?
This week, I finished my Nanodegree in Machine Learning Engineer by Udacity. To finish the course, I had to create a final study. Talking with a dear friend of mine, Pedro, he said that the National Highway Police of Brazil had a dataset on car accidents in federal roads. I decided to study this dataset, and try to predict which types of victims an accident would have based on the local, hour and accident characateristics.
Julia is a programming language that I heard a lot for some time now and I knew it deserved my attention. However, a number of libraries and frameworks for machine learning and deep learning keep emerging and I ended up prioritizing them first.
AKA magics to plug holes in your dataset
When you start entering the data science world, things can become really messy. There are thousands of concepts and meanings, most of them thrown at you at the same time.
This is the second part of my study on predicting the type of victims an accident can have based on the data from the National Highway Police (Polícia Rodoviária Federal), in Brazil. This was my final report for my Machine Learning Engineer Nanodegree and my first technical diploma in the computer science field (yey!).