Coding a Decision Tree from Scratch in Python p.1: Introduction

[@SebastianMantey]

Since there were some questions, the algorithm built in this series is basically the ID3 algorithm. However, instead of using Information Gain I use “Overall Entropy”. To see what the difference is, check out my pinned comment for this video: kzbin.info/www/bejne/hZOvgpamqrF0oqs In case you are interested, I also have a series about coding a random forest from scratch. Here is the link: kzbin.info/www/bejne/jafQgaF9oqiUf68

[@TrueTalenta]

No one compares to you! You are a GURU! No nonsense. Just the business of teaching and to the point. You outshine my PHD professor!

[@SebastianMantey]

Woah, easy easy! I just did a little series on coding a decision tree from scratch. Nonetheless, I appreciate your feedback!

[@nehathakur8221]

been searching for this from days .Best Playlist on Decision tree i have found .thank u sir :)

[@SebastianMantey]

Thanks! Glad it was helpful.

[@anandachetanelikapati6388]

Hats off to you for a clean explanation! My search for clarity on this ends with your Playlist.

[@user-jr1fx3vw2n]

Thank you so much for such a clear demonstration with so many useful details!

[@SebastianMantey]

You are welcome!

[@MrAnandml]

This channel really deserve more subs ... Such a great content ...explained in an elegant manner ...you really got some teaching skills too buddy..

[@SebastianMantey]

Thanks, I appreciate it!

[@minhtriet6873]

Thank you so much, Mantey. You saved my day. All those knowledge i learned from you more than i studied uni, wish all the best for you and your channel will worldwide!!!

[@SebastianMantey]

Happy to hear that!

[@ankurrohilla4655]

just in 5 min i fall in love with ur videos.........thnku very much sir!!!!!!!!!!!!!!!!!

[@adriantang3606]

Dude, your videos have taught me way more than what I learn at uni. Keep up good the work, my guy!

[@SebastianMantey]

Thanks, will do!

[@FloraInn]

Thank you so much for this excellent series! This helps me a lot.

[@halilyaman4498]

Thanks a lot sir ! Your video is my first step to build a complete humanoid robot.

[@coolcolumbus]

Hi, Sebastian. we like your no nonsense to the point presentation. Keep it up.

[@SebastianMantey]

Hi, Neelakanta! Thanks for your nice comments. May I ask you for a favor? I just commented on Siraj’s Q&A video (kzbin.info/www/bejne/kKDOYYeGipV_aq8). Could you maybe like that comment to increase its visibility? The reason I ask is because the last time I had a highly visible comment on one of Siraj’s videos, the video I linked to gained about 600 views. So, your like could potentially be very helpful to me.

[@coolcolumbus]

Yes, sure. Also, try creating a blog page with links to your videos, and share the blog in Facebook etc for more viewers. Good luck.

[@winsonwijaya5592]

new sub! thank you so much for making this content, sir

[@SebastianMantey]

My pleasure!

[@misoandramen]

This is an excellent series, thank you!

[@haiderkamal1106]

BEST TUTORIAL VIDEO SERIES FOR DT !!!! Thank you so much #sebastian Your voice is so soothing...

[@ununbrian]

good video, full support

[@simontimothy6033]

thank you so much! very clear and informative explanation

[@SebastianMantey]

Glad it was helpful!

[@hazemahmed8333]

thank you so much this is pure gold !!

[@petrhanzl8399]

man you just saved my semestr. Thank you :D

[@SebastianMantey]

No problem! By the way, you are not the first one to make such a comment. What’s the assignment that you have to do at college/university?

[@petrhanzl8399]

​@@SebastianMantey Its data processing in python, I chose this data set: www.kaggle.com/sulianova/cardiovascular-disease-dataset/data?fbclid=IwAR2GBHEGq06xmVuYBHwXIDLj7LMWQ6ObXf_bSMl3iSGVeeK3UcZwUe4hc-A My teacher told me to do a decision tree, then agregation. It took me a while to understand the idea of DT. But thanks to you, I can make it.

[@sourabh6496]

Thank you sir you are doing great work🙏

[@viralina8765]

you explained it very well. TQ sir :)

[@mustajabhussain9167]

Thanks for making such an awesome video. Just wanted to ask are you applying the ID3 algorithm in making the decision tree?

[@SebastianMantey]

Thanks! I didn’t specifically wrote the code based on the ID3 algorithm. But I think the diagram, that I based the code on (0:19), is basically the same.

[@nuruddin5136]

@@SebastianMantey actually I didn't get your answer. can you please tell specifically that did you use ID3 ? or you used other algorithms like CART or other. thanks.

[@SebastianMantey]

Hi, Nur! Sorry, for the late reply, but somehow I overlooked your comment in the KZbin studio app. Like I said in the previous comment, I didn’t specifically look at the ID3 or CART algorithms and then tried to implement them in code. What I did was (as far as I can remember): I had a high-level understanding of how decision trees basically work. Then, I tried to explain the algorithm in simple terms and in my own words in these two videos: kzbin.info/www/bejne/jZ2qppywmbuUn7s kzbin.info/www/bejne/hZOvgpamqrF0oqs And then, finally, from the diagram that you can see at 0:19 in this video, I created the code from scratch by myself just using this particular diagram and no other references. I hope this answer makes it more clear.

[@nuruddin5136]

@@SebastianMantey Thanks for your clarification, and it was a nice video to understand.

[@minhletruong2915]

Thanks God I found out you. Thanks so much guy !!!

[@ravindrasah1150]

great video, i loved it

[@tram3634]

Why is it random.seed(0) does not function in googlecolab? I'm getting same indices with or without random.seed(0).

[@SebastianMantey]

That’s strange! When I run the line “random.sample(population=range(100), k=3)” in google colab, then I get different numbers whenever I run the cell. If I put a random seed before that (in the same cell), then I always get the same numbers.

[@exoticme4760]

THank you so much guy!!

[@SebastianMantey]

You're welcome!

[@maximk7816]

Hi! Is there a way do download actual files from Github? I tried to use your link and found the data, but I couldn't download them. So I just made the same file myself using the data. And another question: How you can refer to files directly in Jupiter without path? UPDATE: now I know how to handle files in Jupyter) I'd still like to know how to download files from Github. Anyway, thanks for tutorial! It was great. It was easy to follow along and understandable!

[@SebastianMantey]

Thanks! Here is how to download the files from GitHub: If you click the link to the GitHub repo, then you will see a green button that says “Code”. If you click on that, a pop-up window opens. And there, you can click “Download ZIP” to download the files.

[@tushargoel3521]

can you also provide us this sample data for decision tree

[@crizon222]

I'm entering this playlist to apply the cart algorithm for regression, am I in the right direction?

[@SebastianMantey]

Later on in the series, I am also covering how to use decision trees for regression: kzbin.info/www/bejne/gaXLpqh-h9Zlmqc

[@arvindpc5741]

Nice tutorial. Can you please let me know is this C4.5 or CART ?

[@SebastianMantey]

Actually, I don’t know. What I mean by that is: I don‘t know what the specifics of these algorithms are and how they compare to my algorithm. If you are interested in how I implemented the decision tree, please have a look at the comment of Mustajab Hussain (in the comment section of this video). He asked the same question.

[@merepapa350]

Love ur video

[@brahmlife]

can you have sub features of the features or attributes in the decision tree algorithm?

[@SebastianMantey]

I'm not really sure what you mean by that. Could you elaborate on that?

[@nrted3877]

you look like brad pitt, awseome video

[@ghzich017]

Dataset?

[@prakashdas8518]

Do you have scratch videos on Deep learning

[@SebastianMantey]

Yes! Here is the video with just the code: kzbin.info/www/bejne/hpTJdHiqaqyXgrs And if you want to know all the detailed, theoretical background, you can also watch the whole playlist: kzbin.info/www/bejne/apPZe4V7arBoppY

[@marcelohuerta2727]

is this id3 algorithm?

[@SebastianMantey]

Please have a look at the comment from mustajabhussain9167 for the answer.

[@IvarHilferink]

switch to dark mode powerpoints.