I Survived the Data Science Bootcamp.

Schermafbeelding-2019-05-02-om-14.46.09-1080x675 copy.jpg

This Is What I have Learned.

In the past month, our latest group of Data Scientists worked really hard to take their data science skills to the next level.

After an intensive month of training sessions and hands-on workshops on many different subjects, Aleide, one of the survivors, shares her experiences with her first month of a being a Data Science Xccelerator.

This last group is very mixed and we could potentially learn a lot from each other! But without doubt, most of the knowledge came from our Data Science Lead Matthijs.

Week 1: Winning Games with Algorithms

The first week consisted of the main technical themes. We started with some git collaboration and some python recap for the first day. We learned about the usefulness of implementing software engineering skills, such as creating a command line interface, writing tests with pytest, logging and continuous integration. At the end of the first week, we teamed up and challenged each other by writing the winning algorithm for a game. This was a nice way to incorporate all the new things that we have learned.

Week 2: Dockerizing Web Scrapers

The week after, we build our own web scraper. Matthijs taught us more about databases and frontend development in order to create a Flask API that could show the results of the scraper. We deployed our API on a Google cloud server. By using Cron, we scheduled our scraper to refresh every hour. As a bonus, we Dockerized our web scraper and Flask application. At the end of the week, we teamed up again to solve a complex cipher.

Week 3: Writing Fast and Clean Code

Week three was all about Numpy, Pandas and PySpark! The main focus was to write fast and clean code. While Numpy is all about fast Python, Pandas method chaining helped us with writing clean and clear code. We tested our new skills on multiple datasets. At the end of the week, we were able to tackle a large Twitter dataset and provide a useful hashtag analysis using PySpark.

Week 4: Deploying Models with Sklearn

In the last week of the boot camp, the main focus was Sklearn and machine learning deployment. We learned why Sklearn pipelines and transformers are important to deliver reliable models and why custom metrics can be useful. In the last two days of the bootcamp, we did our assessment again. This was a very nice and fun way to see how much we have learned in the past weeks!

What’s Next?

The next months, we will all be working on a project at our external client, where we are able to strengthen our newly developed skills by applying them to real challenges! Every Thursday, we can be found at the Xccelerated office, either working on our technical skills or soft skills!

Ischa Vieten