Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Activity: Model Evaluation - Activity Description & Resources
This learning activity is so we can learn how to properly evaluate the machine learning models we create as learning activities. You can either create a new model and evaluate it using a method that's appropriate for that model, or you can learn a new evaluation technique and apply it to one of the models you created in a previous activity.

Sebastian Raschka introduced some validation and model evaluation concepts like k-fold cross-validation in the Becoming a Data Scientist Podcast Episode 8 (skip to about 58 minutes), and we reference chapter 6 of his book, Python Machine Learning.

Check out Sebastian's Github repository for the book, which includes code for the chapter (more than is in the book) and demonstrates techniques for several model evaluation metrics.

I'll list more learning resources below. Please add your favorite resources to this thread, and start a new thread to explain which new evaluation techique you used, and how it helped you better understand your model!

"How to Evaluate Machine Learning models" series of Dato blog posts by Alice Zheng

benhammer on github evaluation metrics implementations in several languages

Model selection and evaluation with scikit-learn (python)

Classification evaluation metrics ®

ROC and AUC explained (Data School on YouTube)

Chapter 2 of the "Introduction to Machine Learning with R" class on DataCamp is all about Performance Measures
The Becoming a Data Scientist Podcast Data Science Learning Club is now sponsored by Data CampSee this thread for more info and a coupon. (must be logged-in to view)

Forum Jump:

Users browsing this thread: 1 Guest(s)

About Becoming A Data Scientist is a blog created by Renee Teate to track her path from "SQL Data Analyst pursuing an Engineering Master's Degree" to "Data Scientist". She created this club so participants can work together and help one another learn data science. See her other site DataSciGuide for more learning resources.

Sponsored by DataCamp!