This repository holds example projects that one can see on the DAGWorks platform.
Visit https://www.dagworks.io to sign up!
Here are some Hamilton examples, that use the DAGWorks Driver to log information, that you can subsequently view on the DAGWorks Platform!
Note: you will not have write access for these examples, but if you create your own project, and update the project_id, username, and API key you will be able to write to your own project. Better yet: use the dagworks CLI to generate these for you!
pip install dagworks-sdk
dagworks init \
--api-key API_KEY_HERE \
--username EMAIL_HERE \
--project-id PROJECT_ID_FROM_DAGWORKS_PLATFORM \
--template [hello_world|machine_learning|time_series_feature_engineering|data_processing] \
--location LOCATION_TO_CREATE_PROJECT
Example project that shows the "hello world" project that Hamilton often uses. It's very basic, with one python module that defines a Hamilton dataflow to compute a few pandas columns.
To see this example - navigate here.
Example data processing project. This is a contrived example doing some basic data processing, and how one might structure it with Hamilton and then subsequently view it in the DAGWorks platform.
To see this example - navigate here.
Example machine learning project using the Iris data set. Look at me for a basic ML pipeline set up with Hamilton.
To see this example - navigate here.
Example project showing time-series feature engineering. This is a basic set of transforms based on F33's WaffleML repostiory that uses Hamilton for feature engineering.
To see this example - navigate here.
- Code changes. Structural ones that result in different DAGs, as well as the internals of a function.
- Execution failures.
- Data quality failures.
- Differences in data observability.
- Runs with different configuration.
To see what was changed, see the tags for each version.