hgrif / dbt_tutorial Goto Github PK

Hello!

DBT tutorial

• "live" should be "life"
• semicolons should be commas
• "analytics engineering tool for": remove "for"
• '"enables data analysts ...': should be "enable"

Load data in BigQuery

• CVS and ID are make plural so: CVSes, IDs
• the files are not zipped
• () at end of sentence

Set up development environment

• mention that the env must be python3.6+ to work with dbt
• when i installed dbt via conda, it seemed to not include a number of dependencies of dbt-core; pip installing works, however
• maybe mention you need to change the value of projects
• maybe mention you need to set the value of location to match that of your BigQuery dataset

Load and test data sources

• "...were done uploading ...": should read "...were done via uploading..."
• "...names and fixes any...": shold read "...names and fixing any..."

Transform data

• "...to inspect how these transformations": missing a at the end
• what is the command that causes dbt to touch the files in models/transform/?

Conclusion

• "...extended the Spark": also missing a

general

• remove unused data/ and analysis/ in repo?
• i would move the sentence "The first step of the transformation phase is harmonizing column names and fixes any incorrect types." to the beginning of section "Transform data"
• after the sentence that begins with "The extract and load phase were done...", I would add something like: "But what does dbt run actually do?" and then quickly mention walk through the files (or pieces of files) that dbt run then touches.
• how a brief mention at the beginning of the added value of dbt? what are we using now, why does that suck, and how does dbt try to fix it. as in, you write, 'It aims to "enables data analysts and engineers to transform data in their warehouses more effectively' but you still need to say what "more effectively' means.