Week #4 - General Discussion Thread

[UMGCCCFCSR]

This is the General Discussion post for Week # 4 - Introduction to Tidyverse. What was your favorite of the functions? Which do you think will take you a while to grasp the syntax? Morbidly curious what the approach using base R was before those functions came around? This is thread to have those conversations in.

Cheers!
David

[Rhodnius]

I am just anticipating what will come today and suggest to use read_csv() instead read.csv(). In my experience, read_csv() is more flexible and faster, that will facilitate the whole process when handling several files. Also read_csv() has the additional advantage that is part of the tidyverse.

[UMGCCCFCSR]

@Rhodnius, all fair points, I will add it as a note. I went with read.csv just because I didn't want to have to explain the difference between a data.frame and a tibble this early on 😆

[okolupaev]

At the very end of the text instructions:
"And finally, since I may want to send the data to a supervisor, let’s go ahead and export this “tidyed” version of our data.frame out to it’s own .fcs file. Working within our project folder, this would look like this:"
I think you meant .csv file because the code below generates new .csv file.

Not a big deal but can cause some confusion.

[DavidRach]

Nice catch, will edit it

[KenQuayle]

This is probably a naive question, but what is the difference between running several functions in a long pipe vs running them one at a time to accomplish the same thing? Is it an aesthetic choice or is the long pipe somehow more computationally efficient?

[DavidRach]

Running them as their own line of codes while creating new variables each time through the assignment arrow would also add intermediate objects to your environment, which can clutter up memory. You can however avoid this by overwriting the same object name each individal line.

Main advantage of a pipe in my day to day practice is the ability to troubleshoot. If I highlight a section and run Ctrl + Enter, I can see the intermediate (which wouldn't be the case in the overwrite scenario). I add the next pipe argument over and repeat, I can then see the next intermediate, etc. working down until I get to my final object.

So probably partly convenience, partially aesthetics (readability?). Not about to go proselytize about pipes, but do use them throughout my own workflows.

[MilaAkh]

I just wanted to say thank you!
I haven’t been submitting homework, and I don’t usually appear in the live sessions, so you probably don’t see me. But I am carefully and consistently following the course. So far, I’ve been able to work through the material using the written syllabus.

I’m truly very grateful for your work — this course really matters to me

[DavidRach]

Aww, thanks @MilaAkh!

[claude-chew]

Is there a homework_week4 branch? I didn't see it when creating a pull request.

And to echo @MilaAkh's sentiments, thank you too! This has been a tremendous help and we are only in week 4!

[DavidRach]

Thanks @claude-chew, there is now both a week4 and week5 branch!

[claude-chew]

Thanks @DavidRach !