Building an LLM fine-tuning Dataset
1:01:55
Visualizing Neural Network Internals
53:41
Getting Back on Grid
21:09
10 ай бұрын
INFINITE Inference Power for AI
18:02
Better Attention is All You Need
14:29
Building an Open Assistant API
41:11
ChatGPT API in Python
35:18
Жыл бұрын
Image Editing A.I.
10:39
Жыл бұрын
ChatGPT Writes a Chatbot AI
22:15
Жыл бұрын
Google's DreamFusion AI: Text to 3D
13:46
Open AI’s Whisper is Amazing!
25:51
$5 MILLION AI for FREE
27:25
2 жыл бұрын
Пікірлер
@datajake2742
@datajake2742 15 сағат бұрын
If model.learn returns a trained model (per the docs), what is the for loop doing?
@Ahamshep
@Ahamshep 19 сағат бұрын
In the last few years I followed many of your python videos and your neural networks from scratch series. I disagree that LLM's like chatGPT and claud can replace python content you and others have made. Often times LLM's do provide an overwhelming amount of explanations and examples. Although when your learning something new and errors are produced, you can spend more time on google and stack exchange trying to figure out what is and is not correct. This in itself is not a bad learning exercise although its not structured and its easy to lose interest without clear goals. I hope you do continue producing python content again in the future.
@Stelath
@Stelath Күн бұрын
You're PyTorch explainer videos singlehandedly opened the door for ML stuff for me, I had nearly given up on learning it because I couldn't find anyone who could explain it clearly (at least to someone in high school) and now thanks to those tutorials I'm an ML student researcher with a couple papers under my belt at my university. I agree that LLMs are great for learning but some people are born with a gift for explaining things and you have that gift.
@AlmirAlam.
@AlmirAlam. Күн бұрын
Will part 10 ever come out... 🥲🥲
@AceBurn90
@AceBurn90 2 күн бұрын
1:40 -> Never, I mean NEVER, put the motherboard on top of the plastic packaging it was in to assemble your components. The foil is only antistatic on the inside. Static discharge can still occur on the MB from the outside and damage the components. Simply place the MB on the MB box. That is enough.
@LevitatingSarcophagus
@LevitatingSarcophagus 2 күн бұрын
Hey man, I don't usually comment much but I really wanted to compliment the way you teach. You seem to be having fun while doing what you are doing and it is quite refreshing to see that. Especially when the other tutors sound so bland and monotonous.
@capablancastyle
@capablancastyle 2 күн бұрын
Gracias por tanto y feliz año nuevo!!!
@riccardocarroccio6759
@riccardocarroccio6759 2 күн бұрын
Why is the last output of dimension 3? I don't understand why the size of the X input matrix of initial values should matter at the end. Is it possible to give an example of task so one could relate with the numbers involved in all this?
@AsifHowlader4
@AsifHowlader4 3 күн бұрын
hi sir, can you please explain me, why have you use register on the ground pin ?
@memephiblio
@memephiblio 3 күн бұрын
Cheers man, I'm glad your living well.
@cr0mble
@cr0mble 3 күн бұрын
that hoodie is fire asf
@aSome1
@aSome1 4 күн бұрын
1:15 - the South Park Canadian endless argument haha!
@WojciechŻak
@WojciechŻak 4 күн бұрын
Your mouse even squeaks
@ArashRahmati-n4r
@ArashRahmati-n4r 5 күн бұрын
thank you sentdex. I enjoyed watching this video as well as your PyTorch course. Obviously it got confusing at times and I got help from LLMs. And things worked out. Thankss.
@redgiant2617
@redgiant2617 6 күн бұрын
You don't see 100 6:38 because the range is from 1 to 100 but not including 100. If it was (1, 101) there would be a 100.
@Tom-sq2nt
@Tom-sq2nt 6 күн бұрын
OH MY GOSH, The ReLU explaination is unreal
@Engmelli
@Engmelli 6 күн бұрын
Please continue this series
@rafaeel731
@rafaeel731 6 күн бұрын
Old but gold. You know it's 8 years old if you see Sublime text and not VS Code :D
@Pips_Picks
@Pips_Picks 6 күн бұрын
You are amazing 👍
@vinialves12362
@vinialves12362 7 күн бұрын
This animations are just so useful for understanding. Thank you so much. Which software do you use for that?
@henrymaro3638
@henrymaro3638 7 күн бұрын
Hello, the dog vs cat dataset has been removed.i can not follow the tutorial.can you resent it to me.or give me another link
@halemotors4084
@halemotors4084 8 күн бұрын
Fancied having a go with a PI, bit confusing from a beginners point of view, also not many questions answered!
@namesshiba
@namesshiba 8 күн бұрын
I love this. Thank you so much for making a playlist of this project, you're a life saver! You deserve success.
@MrHaykav
@MrHaykav 9 күн бұрын
Thank you so much
@mritunjay3723
@mritunjay3723 9 күн бұрын
This shell tool is dangerous. Imagine you are running a chatbot and in background it is zipping all your files and sending it to the server.....
@DebuggedBug
@DebuggedBug 9 күн бұрын
How much cost your book if it under 30€ I will buy it and is it also avaible on german (bcuz english is not my first language)
@abhimanyutiwari1637
@abhimanyutiwari1637 9 күн бұрын
i have just one doubt, which is what is actually an input in this case ? what i am thinking is we are actually doing a lot of stuff with the output i guess, without even knowing what is actually the input correct me if i am wrong.
@utkarshashinde4948
@utkarshashinde4948 10 күн бұрын
Thanks a lot ..... for making it so smart and easy
@HabibaAbdallah-m6h
@HabibaAbdallah-m6h 10 күн бұрын
هو ازاي الترجمه عربي
@trxlord09
@trxlord09 10 күн бұрын
Don't know if anyone is ever going to see this, but at 3:37 when he runs the code he gets 2.385 for the result of his third neuron. I copied the code just as is with no changes whatsoever, just run on python 3.12, and somehow got 0.20999999999999996. I've managed to figure out that the code isn't adding in the fourth weighed input with the others through trial and error with manual calculations, but for the life of me I don't understand why since the other two neurons gave the exact same results as in the video.
@trxlord09
@trxlord09 10 күн бұрын
The Numpy code at 15:58 works perfectly pine as well! Super weird black magic stuff
@ArashRahmati-n4r
@ArashRahmati-n4r 11 күн бұрын
13:12 the neural network doesn't really know anything about anything (best sentence) 😀
@İleriDurmaz
@İleriDurmaz 12 күн бұрын
I just discovered your channel with an irrelevant video, bro. You have amazing content. Are you still into algorithmic trading?
@Smillerfox
@Smillerfox 12 күн бұрын
Any idea what to do after learning AI, how to start working for companies projects or any other projects to start making a bit of money.
@elliotkong4519
@elliotkong4519 12 күн бұрын
Does your code change now that Nasdaq acquired Quandl? I'm researching that I need to use an API key from Nasdaq now. Thanks!
@ahmadmudassir93
@ahmadmudassir93 13 күн бұрын
why you are transposing of weights2.T in video P.4 in layer2
@jaxhorn-lq6uv
@jaxhorn-lq6uv 13 күн бұрын
14:50 "It'll be fine." story of all coding ever
@brunomello7499
@brunomello7499 14 күн бұрын
those animations are just GREAT! If I may ask, what tool did you use?
@ct0
@ct0 15 күн бұрын
Also, cool you have bees, I have them on my farm too. Look into single brood chamber/box management. good luck!
@Ni8vik-yh1od
@Ni8vik-yh1od 16 күн бұрын
Thx
@Dr.BananaCraft
@Dr.BananaCraft 16 күн бұрын
2:18 and that's why i prefer to not depend on libraries,it seems like i'm in the right tutorial for me :)
@parmachine470
@parmachine470 16 күн бұрын
changing the state of an entangled qubit changes the state of its partner faster than the speed of light???
@adisheshbalaji8590
@adisheshbalaji8590 17 күн бұрын
i cant use the wiki/googl with quandl
@凯特-s8f
@凯特-s8f 17 күн бұрын
It doesn't work 👎👎👎
@AmitChauhan-bb9yf
@AmitChauhan-bb9yf 18 күн бұрын
Who is watching in 2025
@RAMADANOLADIPO-xw6ry
@RAMADANOLADIPO-xw6ry 3 күн бұрын
2024.5
@rohannagar3041
@rohannagar3041 18 күн бұрын
anyone
@ponpokofamily
@ponpokofamily 19 күн бұрын
This is a big revolution. Sendex 2.0 is here! Thanks for great drone view showing around your beautiful hideout.
@bernardinmoutien2359
@bernardinmoutien2359 20 күн бұрын
Well Here on 14 December 2024 This version of Python brings back memories
@dusklight00
@dusklight00 23 күн бұрын
Thanks sentdex for being part of my childhood
@andrelip1
@andrelip1 23 күн бұрын
Back in the day, you and Siraj had the best hands-on videos about AI. I noticed that you were missed, and now I understand why. Keep going, man.