From the course: Vibe Coding for Data Analysts

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Testing different baseline models

Testing different baseline models

From the course: Vibe Coding for Data Analysts

Testing different baseline models

- Which model gives you the best vibes. Once your data is prepped and ready to go, we can get into the real fun of analytics modeling. As we start modeling though, we should remember that no model is best for every situation. We can use vibe coding to try out a few different models. We're going to start simple with a standard linear model, move to a random forest, and then try an XG boost. You'll notice that I'm working in a brand new script, load our training and testing features and outcome, and then use a linear model, RandomForest, and XG Boost, to predict the number one hit variable. It looks like a great suggestion, so we'll go ahead and bring it into our script and evaluate it. It looks like we're bringing in job lib and pandas, and our assistant has made a great choice in using a logistic regression model. Since our outcome is binary, it is the appropriate choice for this model. It is also used classifiers for a RandomForest and XG boost. We can see that we have an accuracy…

Contents