I’ve been using data.table for around 5 years professionally. Cannot thank you devs enough for the relentless pursuit of speed. It staggers me even to date, 8 years on from this vid, still not many people know about data.table Yet my colleagues complain in case of big data. Most are not even aware R is capable of dealing with data size in the region of, well only limited by RAM GREAT WORK, devs!
@asolisca2 жыл бұрын
The tidymodels framework is simply a work of art. What an amazing job the team at RStudio did with these packages!! They work with each other seamlessly, just like the tidyverse for data wrangling!!
@maddoo232 жыл бұрын
By leaf weight, he means leaf's output, right?
@AkaExcel2 жыл бұрын
Dear Szilard, one thing to note that at 11:40 Comparison of Algorithms, somebody dropped LightGBM algorithm, if you could find original Research Papers it shows that LightGBM accuracy was higher than XGBoost, as i remember
@RealDataScienceUSA2 жыл бұрын
LightGBM is similar in accuracy to XGBoost, in some cases (for some datasets) you can get slightly higher accuracy, for some others slightly less - also depends on tuning etc.
@nobrainsnogains2 жыл бұрын
Thanks Szilard, best wishes from California!
@RealDataScienceUSA2 жыл бұрын
Thank you Arno. All the best.
@kevinstyers48692 жыл бұрын
Functions! I totally agree
@timothywalsh8662 жыл бұрын
I wish I would have heard about data.table before learning dplyr
@ahmed007Jaber2 жыл бұрын
Thank u for this Is there a way to spread a long table in xaringan slides? Suppose u would use the iris dataset and need to spread it across slides so that u can print em on paper
@englianhu2 жыл бұрын
很面善,像大哥阿星的同学 --- 郑国威
@hnagaty3 жыл бұрын
Thanks, really useful and insightful.
@yousif_alyousifi3 жыл бұрын
Dear Dr. Kevin Ushey I have read one of your discussion about resolving of R file is blank and empty when opened. I have written and saved an R file and when I reopened again, the contents of the file are not seen (blank) and when opened it using notepad the only data seen also blank. I have installed the R 3.6.4. Tried to close and open in a new window and new session. nothing works. The file is not empty as the file size is shown as 78KB. Could you please help me to open the attached file? Best regards,
@murongyunhai3 жыл бұрын
legend
@pranav74714 жыл бұрын
The guy is so smart, yet no one can understand what he is saying :( Watch StatQuest it is 100x better - kzbin.info/www/bejne/haWnaaqMlqugbKc
@statswithr6024 жыл бұрын
kzbin.info/www/bejne/hHjUhqqqecudpNk
@user-pg5cu3qh2u4 жыл бұрын
Dirk will teach me Rcpp programming next semester!
@nnuthalapati4 жыл бұрын
This is awesome conversation. I really like the suggestion he gave that frustration is nature when learning something new.
@rashawnhoward5644 жыл бұрын
Thank you! I will take the course.
@sanmarivivier12774 жыл бұрын
Great talk, thank you!
@CarolCCabil4 жыл бұрын
Hi sir new friend is here from philippines im enjoy watching your video & thank you for sharing
@patumusiime4 жыл бұрын
The Dropbox link is broken, from where can I download the datasets
@DeViLTh0rn4 жыл бұрын
The multiple unstacking the df so mind blowing 🤯 🤯🤯, thank you so much!
@AbrahamZamudio4 жыл бұрын
Since when is r-bloggers online???
@DeViLTh0rn4 жыл бұрын
This is so goood! Thank you so much..
@DeViLTh0rn4 жыл бұрын
For my reference 12:39
@1234saikrishna4 жыл бұрын
datasets link is not working. How to download these csv files
@yomolandia4 жыл бұрын
I think the datasetes are from the "nycflights13" package.
@spencerhall60524 жыл бұрын
Actually, these are the Houston 2011 flights dataset, which is the "hflights" package, but you could probably use "nycflights13" just fine.
@dianex.35445 жыл бұрын
Good presentation!
@vinaykrishna19895 жыл бұрын
Is it just me who can't follow his words? and caption is doing poor job as well.
@KumarHemjeet2 жыл бұрын
count me in bro.. :/
@fuhaoda5 жыл бұрын
In the pruning and regularization slide, there should be 1/2 for the first 3 terms, right?
@rltangyt8 ай бұрын
You are right. However, there is some flexibility here. If we view gamma as 2 times the original gamma, the formula is still reasonable. Basically, gamma is a tuning parameter so that 2*gamma is still a reasonable tunning parameter.
@88fjoe5 жыл бұрын
"Chinese is my first language, R is my second language, English is the third!"
@guilhermegoncalvesfernande71685 жыл бұрын
Excellent Class by far! But I just have to point something that I think it was an error. At 26:01 his script to calculate the hour and minutes of the dep variable is: mutate(flights, hour = dep %/% 100, minute = dep %% 100) So... Isn't that wrong? Cause the dep is in minutes and to pass to hours is minutes/60, so the right script should be: mutate(flights, hour = dep %/% 60, minute = dep %% 60) I've not finished the class yet, so I don't know if he'll correct it later, but if someone could just feed back me later, I would be thankful!
@pakpoomtivarkornkit85255 жыл бұрын
After reading the paper and watching the video over and over again, finally I understand how to works.
@tulipking5 жыл бұрын
thx
@datasciencetutor48315 жыл бұрын
Thank you for providing this tutorial for free , please refer to honingds.com/blog/intro-to-dplyr/ for an intro to dplyr
@alexanderher76925 жыл бұрын
as a _real_ coder, I despise ugly simply stuff
@lastnamefirstname88436 жыл бұрын
Bro you gotta zoom out so the camera catches the entire screen, let us do the zooming for ourselves
@zakkyang64766 жыл бұрын
I think the z score part is not correct. Should be like this: planes_z <- flights %>% filter(!is.na(arr_delay)) %>% group_by(plane) %>% filter(n() >30) %>% mutate(z_delay = (arr_delay - mean (arr_delay))/sd(arr_delay)) %>% filter(z_delay >=3) %>% select(plane, z_delay) %>% arrange(desc(z_delay)) View(planes_z)
@WahranRai6 жыл бұрын
Shinese everywhere !!!
@narendrapara6976 жыл бұрын
hii sir ,I have some doubts in shiny r ..how can I contact you sir
@kparag016 жыл бұрын
Father figure in R
@СергейШугарев6 жыл бұрын
XGBoost - the algorithm that wins every competition.
@RavinderRam6 жыл бұрын
hadley wickham is brilliant man
@iqbalhabibiehabibie56896 жыл бұрын
How to show html code inside the Console?
@WmHenry956 жыл бұрын
Understand the value of frustration!
@TheCodeVertigo7 жыл бұрын
link notes/ppt please?
@josephbadana50027 жыл бұрын
This is a great piece of information. josephbadanarivera.com/blog/
@khattami2401937 жыл бұрын
I think Pat was pissed 2:54
@km20527 жыл бұрын
thx
@zizhenwu50877 жыл бұрын
A great man. Think about how many stat phd candidates using MCMC cannot graduate on time without the help from Rcpp.
@dreznik7 жыл бұрын
Excellent interview. I am curious about Max's actual opinion on Partial Least Squares (PLS). He seems not to trust it.
@aniketb57 жыл бұрын
Why would you use display instead of a more convenient option stdout?
@densonsmith27 жыл бұрын
Are the slides available for download?
@CompletelyLegit237 жыл бұрын
speakerdeck.com/datasciencela/jeong-yoon-lee-winning-data-science-competitions-data-science-meetup-oct-2015 I found them here