> why not run it on all the data you can Because more data requires more cleaning and standardization (with more edge cases). It also requires a bigger scale t

> why not run it on all the data you can Because more data requires more cleaning and standardization (with more edge cases). It also requires a bigger scale to obtain and process.