Quantico: Data Wrangling

Quantico offers a wide range of operations that assist in the data science & analytics process. One of those major components is data wrangling and that's what I'll be covering in this article. I classify data wrangling as separate from feature engineering which I'll cover in another article. So keep an eye out for that one later on.

Data wrangling often times begins and ends with SQL but with strong data wrangling packages available it can make sense to shift those tasks to R or Python. Data pipelines can be created more quickly, can be created with more flexibility, can be more dynamic, and can be more easily reused with the storage of functions that are common across projects.

With Quantico, data wrangling makes use of the R data.table package under the hood and a user has a variety of tasks available to them that I categorize into five buckets (listed below). The methods list is certainly not comprehensive. If you would like to see additional methods made available, please make requests on the GitHub repo in the issues section: https://github.com/AdrianAntico/Quantico/issues.

Lastly, after you run your data wrangling operations, feel free to jump to the Code Print output tab to retrieve the code that was used to run the operations. If you don't know data.table this is an easy way to help get up to speed.

Top Level Data Wrangling Categories

• Shrink
• Grow
• Datasets
• Pivot
• Columns

Shrink

1. Aggregate data (8 stats methods available)
2. Subset rows
3. Subset columns
4. Sampling & Stratified Sampling

Grow

1. Join (8 types including rolling)
2. Union

Datasets

1. Partition Data
2. Sort Data
3. Model Data Prep
4. Remove Data

Pivot

1. Cast from long to wide
2. Melt from wide to long

Columns

1. Type Casting
2. Rename Columns
3. Add a Time Trend Column
4. Concatenate Columns