Creating a DVC pipeline
Imagine a simple example of a workflow where a document is printed and then scanned to create a signed PDF document, with DVC managing the dependencies and outputs of each stage.
The print
stage depends on printing instructions outlined in print.sh
and produces the pages
output. The scan
stage depends on instructions in scan.sh
and pages
(output of printer
) and produces a signed.pdf
output.
Your task is to design a DVC pipeline outlining the workflow using the dvc stage add
command. Its syntax is
dvc stage add -n <stage_name> -d <dependency> -o <output> <command>
You can add multiple dependencies and outputs with repeated use of -d
and -o
flags, respectively.
NOTE: DVC has already been initialized in the exercise setup. There is no need to run dvc init
again.
This exercise is part of the course
CI/CD for Machine Learning
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
