The quality of this video is top notch. Your use of manim for the vizualisations is also amazing. I see you don't have many subscribers or views. Mark my words, if you consistently put out content like this, in the next 2 years you will have at least 1 million subscribers. And, Thank you for this video!!
was gonna say it was slow, but it grew on me halfway full code access was a ++ and a instant like! We will be watching your career with gate interest gautam bhai.
Great video, excellent graphics and I especially loved the code explanation. I only wish if you had went into a bit more detail on positional embeddings
![[Live สด] การออกรางวัลสลากกินแบ่งรัฐบาล งวดวันที่ 16 พฤศจิกายน 2567](http://i.ytimg.com/vi/wPd7VFfWHIA/mqdefault.jpg)
![F.HERO Ft. JSPKK x ลำไย ไหทองคำ x M-PEE - ไม่สนิทบิดหมด (Thai Riders Anthem) [Official MV]](http://i.ytimg.com/vi/kPJZ0UihBfk/mqdefault.jpg)
Probably the best and most clear ML/Transformer overview I have ever seen!
I wish every ML video was like that. I'm a visual learner and you're making it very intuitive for me.
this has to be a gem youtube's algorithm just gave to me
The quality of this video is top notch. Your use of manim for the vizualisations is also amazing. I see you don't have many subscribers or views. Mark my words, if you consistently put out content like this, in the next 2 years you will have at least 1 million subscribers. And, Thank you for this video!!
Woah! This is an incredibly good explanation! I have never seen someone explain this so clearly and concisely!
was gonna say it was slow, but it grew on me halfway
full code access was a ++ and a instant like!
We will be watching your career with gate interest gautam bhai.
Geniunely one of the best videos about attention and transformers I've ever seen
clear and concise! loved every bit of it ♥
You deserve to go viral! sadly the topic is so niche
Fantastic work with the graphics. So much easier to understand this way!
Thanks, glad you found it useful!
Nice video. Maybe in The next video you’ll show how exactly did you train your transformers
You’ve earned a sub my friend!
That's the best video I have ever seen! Thank you so much
Tahnk you for the video!
Excellent, very good....
Thanks a lot 😃
Thanks for explaining❤
Keep going, good work, thanks!!!
Glad you found it useful.
Great video, excellent graphics and I especially loved the code explanation.
I only wish if you had went into a bit more detail on positional embeddings
Look out for the next video! Will try to improve on this. Thanks.
great! more implementations from scratch
Dude this is awesome ! How did you get this much knowledge? Your educational background?
Thanks a lot. I'm a PhD student in Scientific Machine Learning.
Amazing video, incredible
great explanation
great video!! thank you for this explanation
Glad you liked it...
Well done video!
great sharing
Manim?
Yes
@@0mean1sigma Only?
Yes
It is not from scratch because you use Pytorch. It's like saying "I am going to implement a web server from scratch" and then importing Django.
I'm sorry if the title is misleading but my idea was to explain and implement the architecture of Transformer NN from scratch.