Craft Beer

James Allen

“Fill with mingled cream and amber,

I will drain that glass again.

Such hilarious visions clamber

Through the chambers of my brain.

Quaintest thoughts — queerest fancies,

Come to life and fade away:

What care I how time advances?

I am drinking ale today.”

- Edgar Allan Poe

Working on a craft beer data set to make some visuals, identify takeaways, run some models and then discuss over a beer.

Executive Summary

• Most popular beer is American IPA

• Average abv of beer is 5.8%

• Most popular beer state is Colorado with 47 Breweries

• Most popular beer city is Portland with 17 breweries

• Average breweries per state is 10

• Average breweries per city is roughly 2

• Random forest is the best performing model with 48% accuracy on test data

Planning

• Initial Thoughts:
  • look at the different beer types
  • how many beers are there?
  • where are the beers brewed?
  • is there a most popular beer type?
  • what state / city are breweries most popular?

Acquire

• download beer and brewery csv files to local computer
• import csv files into jupyter notebook
• merge csv files together into new dataframe

Prepare

• use new dataframe from merged csv files
• check for nulls or missing values
• rename columns for readability
• create sub dataframes for exploration
• create model dataframe for modeling

Explore

• take a look at some beer stats
• plot most popular beers
• look at percentage of popular beers
• look at beer alcohol by volume
• look at breweries by state and city

Modeling

KNN

• Train accuracy: 51%

• Validate accuracy: 38%

• Test accuracy: 36%

Decision Tree

• Train accuracy: 47%

• Validate accuracy: 45%

• Test accuracy: 44%

Random Forest

• Train accuracy: 51%

• Validate accuracy: 49%

• Test accuracy: 48%

Conclusion

• Beer is good
• Drink more beer

Salud

Prost

Cheers

Source

• The beer data was obtained from: Craft Beer Data