padilla410
commented
1 year ago Following up from Gericke's first MR: https://github.com/gcook-usgs/ds-pipelines-targets-1-course-static/pull/1#discussion_r1128903550
I called out the need to reduce duplicative code with a helper function (I remember this being called out in my . I think a language mod could be something like "you can look for opportunities to reduce duplicative code". That being said, we want to be careful to prevent actions like totally revamping components like figures.
from Gericke:
In the pipelines 1 training, it talks about splitting the "my_happy_script.R" file into functions and split into new R files across the fetch, process, and visualize folders. I can see how this is to set up reproducible pipelines but I am unsure if I should be adding code to make it run without the targets package (i.e. reading and writing csv files in each "out" folder and manually connecting the workflows together before I get to the instructions on how to do that with targets). Is the intent to show how you break that apart to manually make a workflow with everything broken down into functions, then bring in targets to link the in and out files? I feel like the activity description could be a touch more explicit on what you are looking the student to do. Either a preface of "this is what you'll do and in the next section...". I have an idea of where you are trying to lead me because I have already hunted ahead and looked at a pipeline for running the QMRs. For someone completely fresh to it, they might find this section confusing on exactly what you want them to do.