It's great to see someone takes this kinda step to make learning more fun, especially on this topic which is one of the hardest of all to understand. Thank you guys for taking this initials it'll really helpful for those programmer who are new in this field.
4:18 More explanation regarding point 2 is necessary: * M's are parameters (like in a function), yes? * specifying m number variables if M's are supplied - not clear what is meant by this. Are m's randomly selected M's? What's the logic behind that? * the best split of these m's - what's meant by a split of these m's? * value of m is being held constant - ??? I think this deserves an extra slide with some visual representation.
very good videos, thank you very much but i have two suggestions to make: 1. the drawing hand is just confusing af when its jumping around and not finishing anything, i can't concentrate on what you're saying with that 2. structure: don't start with advantages/disadvantages, explain the principle first and then go to advantages
I agree with both of those. I was distracted by the hand, and it was odd to learn about the advantages and disadvantages before I knew how it worked. I appreciate the explanations in the video though - I'm just agreeing with the suggestions to make future videos even better.
I think this is great. There are plenty of examples of code and technical online but for fundamental understanding, this works just fine. Good to hear South African this coming from South Africa!
thank you very much for the explanation. I have couple of questions: i) What does it mean "the best split of n is used to split the node"? ii) How do you split in a table with different variables the picture of a finger to apply the random forest algorithm? Thank you very much,
I need some clarification. suppose my dataset contains 10 features and the random forest is formed with variable selection (feature selection i.e 3 features). then should test data contain 10 or 3 features. if it contain 3 feature then it predict the class. if it contain 10 feature how can it predict the class?
While it provides a lot of philosophical implications, this video does not have any details on the technical implementations, not even an outline. What happens with the data? Which hyper-parameters really matter and how to optimally pick them?
any references for our claims? Data overfitting, i've read different things about it, some data or research to back up your claims would be welcome. Also, some links to the type of random forests used in the example fields you give would be helpful.
Bro thank you for your video ! I am using a random forest for a classification learning in time series. Does random forest algorithm take into consideration past values (which is important in time series) ? Should I lag my variables ? Many thanks in advance for you answer !
@@ifusubtomepewdiepiewillgiv1569 I did, this was back in time where I had no clue about machine learning. Now I am into it since 2 years. If you want to make your algo learn about past levels or variations you have to indeed lag your variables. However, the training becomes then a bit tricky because you have to train your algos in a certain way that they just use information that they would have access to at a certain period in time. Traditional K-Fold cross-validation is a fallacy in time series machines learning problems. I don't know how advanced you are in ML but I really recommend you read financial advances in machine learning by De Prado. The first part of the book will show you how to deal with time series.
It looks like it's some sort of automated animation. The hand looks like it's set up to draw paths, write out letters, 'fill' pictures diagonally, and drag items onto the screen. Instead of just being static, it has a little wiggle animation to make it seem more lifelike.
lol ... did you think that all the 'Fast Hand Drawing' videos were actually including actual footage of a hand drawing? :P 90% of them do not contain actual footage.
⭐ Kite is a free AI-powered coding assistant that will help you code faster and smarter. The Kite plugin integrates with all the top editors and IDEs to give you smart completions and documentation while you’re typing. I've been using Kite for a few months and I love it! bit.ly/KITEAUTOCOMPLETE
⭐ If you enjoy my work, Id really appreciate a Coffee😎☕ - augmentedstartups.info/BuyMeCoffee
It's great to see someone takes this kinda step to make learning more fun, especially on this topic which is one of the hardest of all to understand. Thank you guys for taking this initials it'll really helpful for those programmer who are new in this field.
I made a mistake, so it is not GIGI index, it is GINI Index. Apologies for the error.
love GIGI index though >D
Very good explanation of random forest algorithm with impressive visuals
the way of explaining is great...
I'm glad you enjoyed it and it was clear to understand 😁
4:18 More explanation regarding point 2 is necessary:
* M's are parameters (like in a function), yes?
* specifying m number variables if M's are supplied - not clear what is meant by this. Are m's randomly selected M's? What's the logic behind that?
* the best split of these m's - what's meant by a split of these m's?
* value of m is being held constant - ???
I think this deserves an extra slide with some visual representation.
Awesome video for my ML course!
Glad you enjoyed it 😁
What would you like to see next?
very good videos, thank you very much but i have two suggestions to make:
1. the drawing hand is just confusing af when its jumping around and not finishing anything, i can't concentrate on what you're saying with that
2. structure: don't start with advantages/disadvantages, explain the principle first and then go to advantages
I agree with both of those. I was distracted by the hand, and it was odd to learn about the advantages and disadvantages before I knew how it worked. I appreciate the explanations in the video though - I'm just agreeing with the suggestions to make future videos even better.
The drawing hand is such a nuisance. So annoying and distracting.
I agree with both points as well
I think this is great. There are plenty of examples of code and technical online but for fundamental understanding, this works just fine. Good to hear South African this coming from South Africa!
Always glad to get a comment from a fellow South African and can identify the accent :). Glad you enjoying my content. Which part of SA are you from?
@@Augmented_AI as soon as i heard that accent had to check
Proudly South African here, good job augmented startups
Thanks guys 😁 really appreciate it!
Great Video!
thank you very much for the explanation. I have couple of questions:
i) What does it mean "the best split of n is used to split the node"?
ii) How do you split in a table with different variables the picture of a finger to apply the random forest algorithm?
Thank you very much,
what is the difference between this and Multiclass Decision Forest?
Can I ask which software was used to create this animated presentation?
Amazing videos, it is really difficult explaining such deep concepts in a ready for dummies way! keep it up!
Well done man, great content
Shot Bru 😁
Can you give me the name of the software you use for realising these beautiful videos? Thanks in advance
can you please mention which software do you use for making this kinda animated tutorial.
All machine learning tutorials are great!
+Rana Muhammad Kashif thank you so much :).I appreciate the comment.
Hi,
Please correct GIGI to GINI Index
What software are you using for the animations?
just meow !
What's funny is I have a project using random forest in remote sensing on actual forest image classification, like for mapping real trees.
Oh the Irony 🤣👍. Did it work for it?
@@Augmented_AI yeah, it worked well for the large datasets haha
its gini or giri or gigi index approach?
wonderful explanation of a complex but very interesting concept
Your voice and accent are so soothing to listen to. Perfect video! Rate 10/10!
Mejor canal educativo, buen contenido excelente
That is helpful, thanks
the way you explained the difference between boosting and bagging is really vague , which may confuse audience.
you are right, i am lost on this point.
like the quality and visuals in the video
i wish all my lecture are like this! good works!
As far as i know decision tree can not handle missing values , then how come random forest does that?
check surrogate splits
I need some clarification. suppose my dataset contains 10 features and the random forest is formed with variable selection (feature selection i.e 3 features). then should test data contain 10 or 3 features. if it contain 3 feature then it predict the class. if it contain 10 feature how can it predict the class?
Thanks
I thank for great explanation and i have one doubt in case of 50% how rf will be work?
sir what should i refer to study detailed algorithm ... any book or journal
the best, Really great explanation.
+zaibi Ejaz thank you, I really appreciate it :)
I keep pausing the videos to check the source of the music audio until I realized it is the backsound haha
Clear and funny explanation! Thanks!
Thanks for sharing , as far as concern about RF has no pruning due to mtry. I think there is max_depth parameter which limit to further grow of trees?
Thanks !! Excellent Job!
While it provides a lot of philosophical implications, this video does not have any details on the technical implementations, not even an outline. What happens with the data? Which hyper-parameters really matter and how to optimally pick them?
any references for our claims?
Data overfitting, i've read different things about it, some data or research to back up your claims would be welcome.
Also, some links to the type of random forests used in the example fields you give would be helpful.
Great series, easy to understand, except the index - is it Gigi, Giri or should it be Gini - Corrado Gini the inventor of the index.
Hi Robert. Yeah I made a lil typo there :p. But thanks 😎, glad you enjoyed the series.
Alguien podria traducir este gran video por favor ya que el google no realiza bien la traduccion saludos
0.41
GIGI index
Gini index
Thanks for the video!
This really helpful .. clear my concept of random forest
good explanation and presentation - thanks!
Bro thank you for your video !
I am using a random forest for a classification learning in time series.
Does random forest algorithm take into consideration past values (which is important in time series) ? Should I lag my variables ?
Many thanks in advance for you answer !
did u ever find out
@@ifusubtomepewdiepiewillgiv1569 I did, this was back in time where I had no clue about machine learning. Now I am into it since 2 years.
If you want to make your algo learn about past levels or variations you have to indeed lag your variables.
However, the training becomes then a bit tricky because you have to train your algos in a certain way that they just use information that they would have access to at a certain period in time. Traditional K-Fold cross-validation is a fallacy in time series machines learning problems.
I don't know how advanced you are in ML but I really recommend you read financial advances in machine learning by De Prado. The first part of the book will show you how to deal with time series.
thanks the video helped a lot
😊 Thanks for sharing! ✌ 👏🏻
You are most welcome :)
at 2:03, you seem to say that random forest perform better at classification than regression, while you wrote the opposite. Or did I miss something?
Hi, could you please tell me how you edit these video to create this intersting effect?
+张荣恩 I use sparkol videoscribe
gr8 work .......
gr8 explanations...
+Mak S thank you , I really appreciate it :)
why is the hand moving when it is not actually drawing anything?
It looks like it's some sort of automated animation. The hand looks like it's set up to draw paths, write out letters, 'fill' pictures diagonally, and drag items onto the screen. Instead of just being static, it has a little wiggle animation to make it seem more lifelike.
lol ... did you think that all the 'Fast Hand Drawing' videos were actually including actual footage of a hand drawing? :P 90% of them do not contain actual footage.
Yeah, this is one of the most distracting drawing animations I've ever seen.
Why did you censor the male though 🤔
😁😆
Cool, but stop adding that hand there.
STFU
@@samram2950 it's a waste of time and information
It is GINI Index approach
correct, at ~0:54
good video
I'm glad you enjoyed it 😊
Best fckin random forest video in TH-cam.
TH-cam want to play that 0:10 1/2 ads game? Took 6x longer to write a comment but paused the ads and skipped the vid. Fuck you youtube. Respectfully
Summarized really well ! #RandomForest
That's awesome!
Dude your accent is so South African!
I am South African 😁
The writing hand is an unnecessary distraction.
So it's basically a democracy
true while learn()
⭐ Kite is a free AI-powered coding assistant that will help you code faster and smarter. The Kite plugin integrates with all the top editors and IDEs to give you smart completions and documentation while you’re typing. I've been using Kite for a few months and I love it! bit.ly/KITEAUTOCOMPLETE
distracted by lip smacking noise, maybe some water would help
Thank you for the suggestion. Will try that and edit it I post.
I didn't understand the "VOTE WOMEN" part.
5:16 Lol
PARLI MALISSIMO IN INGLESE
Nicely done, however nothing was explained. Neither math nor implementation which makes this mostly useful for a very first introduction
didnt understand shit about pseudo code no explanation only textbook reading
very hard to understand
First!!!!
its gini or giri or gigi index approach?