Thank you for posting this. Alot of people nowdays think deep learning is import model, import data run and voila (fast.ai course). A good analogy I read somewhere was like driving a car. You don't have to know ins and outs of a car to drive a car, which is fine if you just want to drive. But if you want to be a mechanic, you better know the ins and outs. When things go wrong, and they will go wrong, you won't even know where to begin fixing it if you don't understand the fundamentals behind it. Also, once you feel comfortable, Kaggle is a fun way to test your abilities. Initially, your goal shouldn't be to win competitions, just get a grounding on how to structure your project and get some real life problems to play with.
Awesome :) Yeah it can be tough to know where to start learning and in the beginning it can just feel overwhelming. Wish you the best of journey in deep learning!
Really grateful for this video, I've currently got a background based entirely on finance and economics with no prior knowledge of programming. This video has been extremely helpful in deciding where to start. Many thanks for the great content and the proffesionalism shown.
1. Learn linear algebra resources: khan academy, MIT 2. Basic programming (python) 3. Basics of machine learning: coursera Andrew Ng. 4. Deep Learning specializayion: coursera Andrew Ng
Heya Aladdin! Thank you so much for this effort. I'm so glad there is an option where people take the time to go through all of those research papers word-by-word (almost) and explain them in a way someone like me can understand. The hardest part for me is trying to relate what I learned; not that I am able to understand anything much, and make some connections that'll make sense of things. Seeing them materializing by example from scratch helps a lot. I am a firm believer in maturing comprehension before making a leap into projects. Please keep at it with this format. It's amazing and I believe, if I ever become just as good as you are, I'd make a paper walkthrough too to help people make sense of what they are reading.
Definitely, I think more people should document and share what they are doing. Doesn't matter what stage you're at, if you're just learning then share that too. In my eyes I'm still a beginner:)
Thanks a lot man ! i have just finished the A.i Specialization course, and this video gave me a nice vision for what to do and how to advance in the A.i journey .❤
You are true about about calculas being the second most important and so is Andrew Ng. But the difference is you are giving the order for Deep Learning and he gave for Machine Learning.
Your video is very informative. Kudos to your efforts!!! Keep doing more videos and share your learning methods, it will be very useful for general people to enter into AI world. Thanks a lot and wish you success and abundance!!!
Do you know I watched this video now for first time , I learnt machine learning by Anrew Ng , deep learning specialization by andrew Ng , and now I am also learning CS231n and simultaneously doing kaggle compitions and building my own models, before watching this video I didn't know that what I am doing is actually a prescribed way of learning by you, before watching this video I was wondering whether I am learning right things or not or I am just wasting my time in courses 😔 but you gave me a motivation that whatever I learnt is right and whatever I am learning is also right .... thanks bro for your indirect support it really encouraged me🙏🥳🥳
Hi Aladdin, Thanks for the inputs. You can also discuss youtube channels related to machine learning and deep learning that will be helpful for the beginners
Thank you so much for this wonderful video! With all the recent exploding advances in AI these past few months, it has made me interested in AI. I have a software engineering background and a couple of my favorite languages are C++ as well as Java. Although I am sure it wouldn't be very difficult for me to learn Python -- I know about 20 languages -- I am wondering if I even need to ? Can I do all these courses in Java and or C++? I like using a more typed language.
I am a mathematics PhD student and I have keen interest in ML and AI. Suggest me some good books which explain the actual mathematics behind ML algorithms.
Same position as you. The field is new enough that there aren't really any good books. You should talk to professors in your department who work in ML research and they can tell you some of the folklore.
I need ur advice in something What is the most efficient way in studying these courses ? I'm writing almost 90% of the knowledge i learn in my notebook 📒 , but this way takes really a very long time to finish a course and if i didn't write what i learn , I feel like i didn't understand that point or I'll forget it if i didn't write it. !!!
Thanks for the explanation, I'm at the stage of taking Andrew Ng Deep learning specialisation, but between tensorflow and pytorch which one should i master or i can learn both and used it when the need arise.
I think I will make a video about this because it's something I remember thinking about a lot too. But it really doesn't matter, for example I did the specialization (which is in tensorflow) but then learned more about PyTorch. I think which one you choose will not matter much at all, and long term you will be best off if you know both
hey aladdin , recently i had my college classes and had to write notes of it so i thought of and idea about making a notes generator using deep learning, taking input as images of my classmates notes and generating notes in own style of handwriting. can you tell me how i can go about the idea , how to get the data and how to get stuff working , it would be really helpful.
That sounds very interesting! I haven't done any considerable research into this but what you would need to do is find some suitable dataset, just googling there was a couple and I thought this: www.fki.inf.unibe.ch/databases/iam-handwriting-database Looked interesting. Do some research on what datasets can be used and find as much data as you can from different sources. That would be the initial step in my view and then I would take a pretrained ResNet and train the entire network on the data you collected. Before collecting your own data I would first check if the performance is good on your handwritten notes, and if it's not doing too well on your notes then you would need to get more data and label it for your specific handwriting and then finetune the network on those or something similar to that.
@@thecros1076 come on bro, he already gave you a valuable answer. It's time to start figuring out the rest yourself! Gotta learn how to solve problems 🤷♂️
I think the limiting factor in most cases is about deep learning itself but for learning Pytorch specifically reading the official tutorials, doing projects, searching docs and pytorch forum when you get stuck, and watching youtube videos
Thank you for posting this. Alot of people nowdays think deep learning is import model, import data run and voila (fast.ai course). A good analogy I read somewhere was like driving a car. You don't have to know ins and outs of a car to drive a car, which is fine if you just want to drive. But if you want to be a mechanic, you better know the ins and outs.
When things go wrong, and they will go wrong, you won't even know where to begin fixing it if you don't understand the fundamentals behind it.
Also, once you feel comfortable, Kaggle is a fun way to test your abilities. Initially, your goal shouldn't be to win competitions, just get a grounding on how to structure your project and get some real life problems to play with.
@@AladdinPersson what do you think about the kaggle mini courses? I did both their machine learning courses and learned quite a bit.
@@CE-vd2px I haven't done them but will check them out!
I think I recovered my interest in this area just by having a way to start walking through... thanks
Awesome :) Yeah it can be tough to know where to start learning and in the beginning it can just feel overwhelming. Wish you the best of journey in deep learning!
Dont stop making videos you have a bright future
One of the underrated channel! You deserve more recognition, mate!
Really grateful for this video, I've currently got a background based entirely on finance and economics with no prior knowledge of programming. This video has been extremely helpful in deciding where to start. Many thanks for the great content and the proffesionalism shown.
I really appreciate the kind words :)
Replicating research paper sounds great🔥 could you please talk about more on how to go approach this way of learning.
1. Learn linear algebra
resources: khan academy, MIT
2. Basic programming (python)
3. Basics of machine learning: coursera Andrew Ng.
4. Deep Learning specializayion: coursera Andrew Ng
Heya Aladdin! Thank you so much for this effort. I'm so glad there is an option where people take the time to go through all of those research papers word-by-word (almost) and explain them in a way someone like me can understand. The hardest part for me is trying to relate what I learned; not that I am able to understand anything much, and make some connections that'll make sense of things. Seeing them materializing by example from scratch helps a lot. I am a firm believer in maturing comprehension before making a leap into projects. Please keep at it with this format. It's amazing and I believe, if I ever become just as good as you are, I'd make a paper walkthrough too to help people make sense of what they are reading.
Definitely, I think more people should document and share what they are doing. Doesn't matter what stage you're at, if you're just learning then share that too. In my eyes I'm still a beginner:)
Thanks a lot man !
i have just finished the A.i Specialization course, and this video gave me a nice vision for what to do and how to advance in the A.i journey .❤
Man you can't imagine how much this is helping me ! Thank you thank you thank you
Oh damn, thanks to you I learned that Fridman had a podcast with Andrew Ng. Jumping to listen to it right away!
Can you also make a video about kaggle, how should I integrate it with my Journey in Deep learning
WAW! You are so young and so smart,congratulations!
I enjoyed watching this, thanks for some ideas you've sparked ;) Subscribed and liked
❤️
Great job keep going bro ....I'm from Algeria i love your courses
Thanks Aladdin!!
make one video on how to get jobs in deep learning or computer vision ,it will clear all doubt of being this method efficient and helpful
You are true about about calculas being the second most important and so is Andrew Ng. But the difference is you are giving the order for Deep Learning and he gave for Machine Learning.
Your video is very informative. Kudos to your efforts!!! Keep doing more videos and share your learning methods, it will be very useful for general people to enter into AI world. Thanks a lot and wish you success and abundance!!!
Really appreciate the kind comment
Thanks for sharing!
you are awesome man..! you are the best teacher I have ever had
Do you know I watched this video now for first time , I learnt machine learning by Anrew Ng , deep learning specialization by andrew Ng , and now I am also learning CS231n and simultaneously doing kaggle compitions and building my own models, before watching this video I didn't know that what I am doing is actually a prescribed way of learning by you, before watching this video I was wondering whether I am learning right things or not or I am just wasting my time in courses 😔 but you gave me a motivation that whatever I learnt is right and whatever I am learning is also right ....
thanks bro for your indirect support it really encouraged me🙏🥳🥳
You really had to put the cross entropy formula in the intro.
Hi Aladdin,
Thanks for the inputs. You can also discuss youtube channels related to machine learning and deep learning that will be helpful for the beginners
such a dreamy eyes and voice (very weird how deep learning in my head work)
Thanks a lot, useful informations and enjoyable video.
linear algebra
probability and statistics
calculus ( including multivariate calculus)
optimization
This video motivate me a lot. Thank You Aladdin
So yeah I'm wrong and he's right hahah love the humility.
greet content we need ,thanks
Hey Aladdin. Do you take notes while taking these courses? If yes, where?
Your videos are really helpful.
Thank you!
Great video
Thank you so much for this video. Which course or book do you recommend for statistics and for features engineering?
I don't know, I'm not an expert in those areas. I need book recommendations on this also:)
Thank you
Thank you so much for this wonderful video! With all the recent exploding advances in AI these past few months, it has made me interested in AI. I have a software engineering background and a couple of my favorite languages are C++ as well as Java. Although I am sure it wouldn't be very difficult for me to learn Python -- I know about 20 languages -- I am wondering if I even need to ? Can I do all these courses in Java and or C++? I like using a more typed language.
Python is used most in ai programming
Coursera courses are priced differently in different countries. I'm from India, so If I take a 6 month subscription, it only costs me $229 in total.
I am a mathematics PhD student and I have keen interest in ML and AI. Suggest me some good books which explain the actual mathematics behind ML algorithms.
Same position as you. The field is new enough that there aren't really any good books. You should talk to professors in your department who work in ML research and they can tell you some of the folklore.
Make a video on PyTorch lightning too
Will look into it, right now I don't have much experience with PyTorch lightning
People : HOW TO LEARN DEEP LEARNING
Aladdin : Andrew Ng
Great video. I'm almost done Andrew NG's course, and would really like to work upon CNNs. What kind of research papers should I replicate?
great
Should I learn SQL?
I need ur advice in something
What is the most efficient way in studying these courses ?
I'm writing almost 90% of the knowledge i learn in my notebook 📒 , but this way takes really a very long time to finish a course and if i didn't write what i learn , I feel like i didn't understand that point or I'll forget it if i didn't write it. !!!
Keep making videos
Can you do a tutorial on word2vec from scratch ?
I'm currently a bit busy doing TensorFlow tutorials but will look into this after! :)
Thanks for the explanation, I'm at the stage of taking Andrew Ng Deep learning specialisation, but between tensorflow and pytorch which one should i master or i can learn both and used it when the need arise.
I think I will make a video about this because it's something I remember thinking about a lot too. But it really doesn't matter, for example I did the specialization (which is in tensorflow) but then learned more about PyTorch. I think which one you choose will not matter much at all, and long term you will be best off if you know both
@@AladdinPersson learn fast.ai
hey aladdin ,
recently i had my college classes and had to write notes of it so i thought of and idea about making a notes generator using deep learning, taking input as images of my classmates notes and generating notes in own style of handwriting. can you tell me how i can go about the idea , how to get the data and how to get stuff working , it would be really helpful.
That sounds very interesting! I haven't done any considerable research into this but what you would need to do is find some suitable dataset, just googling there was a couple and I thought this: www.fki.inf.unibe.ch/databases/iam-handwriting-database
Looked interesting. Do some research on what datasets can be used and find as much data as you can from different sources. That would be the initial step in my view and then I would take a pretrained ResNet and train the entire network on the data you collected. Before collecting your own data I would first check if the performance is good on your handwritten notes, and if it's not doing too well on your notes then you would need to get more data and label it for your specific handwriting and then finetune the network on those or something similar to that.
@@AladdinPersson thank you so much ....but I hope you can make a video about this project😬😬?
@@thecros1076 come on bro, he already gave you a valuable answer. It's time to start figuring out the rest yourself! Gotta learn how to solve problems 🤷♂️
Why is python the standard language for deep learning
What about Pytorch? Any tips?
I think the limiting factor in most cases is about deep learning itself but for learning Pytorch specifically reading the official tutorials, doing projects, searching docs and pytorch forum when you get stuck, and watching youtube videos
How much math do we have to learn brother because as u know math is too much
4:50 yeah that's right lol
首先，感谢您分享您的经验，但我觉得每次观看视频时我都会感觉到。事实上，我似乎什么都没学到。我很苦恼。你有什么解决方案吗？
@4:01 ..... you can find it for free , idk LOL
he knows
math is my game
but I don't know any thing about Algorithm and data structures
You know you should change the title to how to teach people to learn deep learning. It really is misleading.