Nice introduction. I remember starting with R and hating it compared to Python. But it is quite prevalent in Bioinformatics. Luckily after a while, the Stockholm Syndrom kicked in and I started to like it 🤣
Awesome video. As a very beginner, I found your explanation very clear, enough to find my feet in the process. One question: How does the tidyverse cluster combine with other larger packages? I am specifically interested in a sequencing data package such as DSeq2, haven't really seen others use any of these packages with such very niche pipelines.
great video!! personally I would greatly benefit from learning more about the purrr package and list-columns, especially in the use case you had mentioned regarding simulations. I would love a video on that!
@very-normal Let me try to answer this question, but first let me start with a disclaimer: You know R much better than I do (so thanks for the excellent overview!), so all with a grain of salt here: 1. Readability matters. When your interpreter interprets your if statement, it reads something like this: IF (condition is met) DO {something}, ELSE DO {nothing} And for convenience reasons, this ELSE DO {nothing} is left away. But if you want the program to do something, adding an else statement is just making explicit, what you implicitly wrote anyway. 2. I mistrust code without defaults. If adding an extra else statement causes problems, maybe there is something wrong with my code in the first place. Maybe I haven't thought about what values my condition can evaluate to. It doesn't hurt to throw an error when the condition evaluates to some non-sense that shouldn't happen. (I like Rust mentality: write robust programs, where everything is accounted for.) Too often R code is written on the fly and people spend hours debugging some weird results. Maybe the R interpreter and paradigm are to blame for some of this, but I believe writing robust code is a virtue in itself, even if you are "just doing statistics". Anyways, thanks for the great video ... as always actually. Keep up the great work! 👍
Yeah, I see where you’re coming from! I would be lying if I said that I haven’t written errant R code that’s burned me in the long run. It’s an oversight in graduate statistics programs that good programming conventions are not taught to us. I’ve tried learning by myself, but I definitely know I have a lot to learn here. I appreciate your thoughts on it. I know that I’ve adjusted the way I code to work with not using defaults, but I could reconsider this. Thanks!
@@very-normalThat's cool, thanks for the considerate reply! Just for the record I want to recommend two books on the matter of good programming style: 1. The Pragmatic Programmer by Andy Hunt and Dave Thomas, which is THE seminal book for best practices in programming, and 2. Advanced R by Hadley Wickham (available for free online), which tries to provide the same, but specifically for the R programming language. I must admit I have only read some of the second recommendation, but it seems pretty good. Oh and then there is Joel Grus who is a serious data scientist with what I find impeccable coding style ... but he's a Python guy (sorry) :P
Nice introduction. I remember starting with R and hating it compared to Python. But it is quite prevalent in Bioinformatics. Luckily after a while, the Stockholm Syndrom kicked in and I started to like it 🤣
Python video would be great! Thanks! 😊
I have a masters in Data Science, and have been using R since 2018. However, today I learned I could NAME list objects :D
Awesome video. As a very beginner, I found your explanation very clear, enough to find my feet in the process.
One question: How does the tidyverse cluster combine with other larger packages? I am specifically interested in a sequencing data package such as DSeq2, haven't really seen others use any of these packages with such very niche pipelines.
Very good organization and explanation in this video.
Python content please 0:41. Good vid
Entertaining, educational, and clear. I like the video can help data scientists at entry levels.
great video!! personally I would greatly benefit from learning more about the purrr package and list-columns, especially in the use case you had mentioned regarding simulations. I would love a video on that!
Thanks for watching! Luckily that video’s been planned out for the month, keep your eyes peeled!
awesome!! Happy New Year!! @@very-normal
Love your videos❤❤❤
And thank you for brilliant explanations ❤🎉😊
Beautiful work
Nice video
Edit: haha maybe that was the point
At 2:38, the comment after array() is pretty similar contrast so it's hard to read.
🔥🔥, python video please
i am using R if i am feels like spss had magical result.
I still prefer you to use R for more statistics stuff in your channel because they are more rigorous to mathematics and statistics.
this was pretty great :)
Please make Python videos too.
5:16 You don't recommend adding an else to an if statement because it's easy to make mistakes? That bugs me. Am I alone on this? 😅
lol why does it bug you?
@very-normal Let me try to answer this question, but first let me start with a disclaimer: You know R much better than I do (so thanks for the excellent overview!), so all with a grain of salt here:
1. Readability matters.
When your interpreter interprets your if statement, it reads something like this:
IF (condition is met) DO {something}, ELSE DO {nothing}
And for convenience reasons, this ELSE DO {nothing} is left away. But if you want the program to do something, adding an else statement is just making explicit, what you implicitly wrote anyway.
2. I mistrust code without defaults.
If adding an extra else statement causes problems, maybe there is something wrong with my code in the first place. Maybe I haven't thought about what values my condition can evaluate to. It doesn't hurt to throw an error when the condition evaluates to some non-sense that shouldn't happen. (I like Rust mentality: write robust programs, where everything is accounted for.) Too often R code is written on the fly and people spend hours debugging some weird results. Maybe the R interpreter and paradigm are to blame for some of this, but I believe writing robust code is a virtue in itself, even if you are "just doing statistics".
Anyways, thanks for the great video ... as always actually.
Keep up the great work! 👍
Yeah, I see where you’re coming from! I would be lying if I said that I haven’t written errant R code that’s burned me in the long run. It’s an oversight in graduate statistics programs that good programming conventions are not taught to us. I’ve tried learning by myself, but I definitely know I have a lot to learn here. I appreciate your thoughts on it. I know that I’ve adjusted the way I code to work with not using defaults, but I could reconsider this. Thanks!
@@very-normalThat's cool, thanks for the considerate reply!
Just for the record I want to recommend two books on the matter of good programming style:
1. The Pragmatic Programmer by Andy Hunt and Dave Thomas, which is THE seminal book for best practices in programming, and
2. Advanced R by Hadley Wickham (available for free online), which tries to provide the same, but specifically for the R programming language.
I must admit I have only read some of the second recommendation, but it seems pretty good.
Oh and then there is Joel Grus who is a serious data scientist with what I find impeccable coding style ... but he's a Python guy (sorry) :P
Ooh, very nice, I’ve seen Advanced R in passing, but never got a look at it. I’ll take a look at all your suggestions! Thanks!