Felipe.png

 

Dr Felipe Aguirre Martinez

Dr Felipe Aguirre Martinez is the Lead Data Scientist at WorkIT Software, Paris. Felipe is Colombian. He is a Mechanical Engineer with a Ph.D. in computer science from the Université de Technologie de Compiègne. Felipe is an alumnus of Data Science Retreat (Batch 09). During his Ph.D. and his first years of professional experience, he worked in the field of reliability and risk analysis. Throughout these years he learned probability, statistics, uncertainty theories in general, Monte Carlo methods, and machine learning basics. Since 2016 he has been working as a Lead Data Scientist at Workit Software, where he is building models that create order out of the untidiness of the web. He has a refactoring monkey living inside his head and he is obsessed with proper coding practices.
https://www.linkedin.com/in/felipeam/
https://github.com/felipeam86/
 

Software development skills on Python for Data Scientists

outcome

By the end of this course, you will feel more confident to develop your portfolio project and iterate through your ideas. You will understand that a proper code structure will give you more freedom on your exploratory analysis and speed up the consolidation of ideas

You will know how to do feature engineering, fit a random forest, you have cracked MNIST with Keras or tensorflow, but you are still locked down to a jupyter notebook. Your are lost on over 1000 lines of code living on the same script.
We will go through basic software development skills for Data Scientists that will help you build a proper Python code structure. Clean and well structured code is easier to maintain, modify and share with others. We will talk about how to separate concerns between data acquisition, feature engineering, model fitting, model prediction and visualization.

Content 

In this course we will go through basic software development skills for Data Scientists that will help you build a proper Python code structure. Clean and well structure code is easier to maintain, modify and share with others. We will talk about how to separate concerns between data acquisition, feature engineering, model fitting, model prediction and visualization.

Additionally, we will see how to do some basic documentation, collaborate on git, efficiently use IDEs like PyCharm or Sublime and how to build REST APIs to serve your predictive models.

PREREQUISITES

  • Basic to intermediate knowledge of machine learning and Python