Excellent timing on the video! I started a Data n Drinks club, and we are trying to use kmeans clustering to fix Ohio's district map. We hope that by clustering based on a set of criteria we can address some of the election problems in the state :)
Julia, thank you for your presentation, Please how do you mange to automatically adjust your scripts positions, like at 6:39 and 10:00 min of presentation?
I am using the keyboard to reindent lines, which is Cmd + I on a Mac, so first I select my whole script (Cmd + A) and then I reindent. You can find keyboard shortcuts here: support.rstudio.com/hc/en-us/articles/200711853-Keyboard-Shortcuts
You put the function in a named list and it will append the function name to the original column. You can adjust how that works with the .names argument
@@UCKszbcV The rainbow parentheses came as an option in RStudio 1.4; you do need to turn it on the option: blog.rstudio.com/2020/11/04/rstudio-1-4-preview-rainbow-parentheses/
If you're interested in time series forecasting with tidymodels, you can check out modeltime from Matt Dancho and some of his videos: business-science.github.io/modeltime/
@@JuliaSilge I know the work of Matt, but I just wonder if there is some best practices for time series forecasting using tidymodels that do not involve additional packages but thanks for the suggestion.
Thanks Julia - that was a really helpful video! Just wondering, are there any other algorithms you recommend as an alternative for cases (like this one) where you don't see clear seperation between clusters?
Hi Julia, brilliant tutorial as always!!! I want to try this technique soon. Also, I was wondering, for STM, you can print out regression coefficients for your topics, would you know if you can manipulate code so that it prints more than 11 lines of coefficients, please? I’ve had it open on StackOverflow: stackoverflow.com/questions/66352787/can-you-print-more-than-11-covariates-for-summary-estimateeffect
Excellent timing on the video! I started a Data n Drinks club, and we are trying to use kmeans clustering to fix Ohio's district map. We hope that by clustering based on a set of criteria we can address some of the election problems in the state :)
Thank you! This is an excellent tutorial on Kmean. Many functions are explored very elegantly.
This channel is a gem!
You nicely enrich the unsupervised literature in UA-cam. Keep going!
Julia, thank you for your presentation, Please how do you mange to automatically adjust your scripts positions, like at 6:39 and 10:00 min of presentation?
I am using the keyboard to reindent lines, which is Cmd + I on a Mac, so first I select my whole script (Cmd + A) and then I reindent. You can find keyboard shortcuts here: support.rstudio.com/hc/en-us/articles/200711853-Keyboard-Shortcuts
Great video, thanks for sharing
love your videos Julia!!! thx a lot!!
Can you pls explain use case of k-mean cluster in security domain
Thanks for the tutorial, the tidy() function is very handy.
Thank you for the great video! Can you please explain how users can then find the centroid for each cluster? Or isolate which cluster is the largest?
The `tidy()` method gives you the centers of the clusters, as well as the size (i.e. how many members there are) for each cluster.
Thank Julia. Awesome videos.
At 8:55 you made some proportions by dividing by the total. How could you append new columns instead of overwriting the originals?
You put the function in a named list and it will append the function name to the original column. You can adjust how that works with the .names argument
Thanks again for the great videos.
Amazing content as always. Thank you!
can u please explain what to_sbake_case do. thanks, why not just to lower case
It converts strings into "snake case" format. en.wikipedia.org/wiki/Snake_case
Thanks @julia silge. What RStudio Theme are you using? :)
It's one of the ones available in the rsthemes package, I believe Oceanic Plus:
www.garrickadenbuie.com/project/rsthemes/
@@JuliaSilge Thanks! Is the different colour of brackets according to their hiercharchy part of theme or something that configure separately?
@@UCKszbcV The rainbow parentheses came as an option in RStudio 1.4; you do need to turn it on the option:
blog.rstudio.com/2020/11/04/rstudio-1-4-preview-rainbow-parentheses/
@@JuliaSilge Thanks for taking the time to answer, you are a ⭐!
your scripts are always so fancy.
Excellent 🙏
Nice video. How about to show some best practices to use tidy models with time series forecasting? You haven’t touch time series in any of your video
If you're interested in time series forecasting with tidymodels, you can check out modeltime from Matt Dancho and some of his videos:
business-science.github.io/modeltime/
@@JuliaSilge I know the work of Matt, but I just wonder if there is some best practices for time series forecasting using tidymodels that do not involve additional packages but thanks for the suggestion.
Thanks Julia - that was a really helpful video! Just wondering, are there any other algorithms you recommend as an alternative for cases (like this one) where you don't see clear seperation between clusters?
I think I would go to something like PCA, UMAP, and friends.
Thank you!
Hi Julia, brilliant tutorial as always!!! I want to try this technique soon.
Also, I was wondering, for STM, you can print out regression coefficients for your topics, would you know if you can manipulate code so that it prints more than 11 lines of coefficients, please?
I’ve had it open on StackOverflow:
stackoverflow.com/questions/66352787/can-you-print-more-than-11-covariates-for-summary-estimateeffect