Corrections and ammendments: 1) At 2:31, 6 mil * 4 mil will be a huge number in trillions. I put 24 million which clearly is a typo. 2) Convolution operation uses only the sum and not the average (that I have indicated in the video at 5:44. At 5:44 the correct value will be -1 and not -0.11) 3) In academia and popular literature the thing they refer as a convolution operation is actually a cross correlation. You can read more about it online but in AI world people call it convolution only so its ok to use that term. But just be aware that looking at it from purely mathematical standpoint, it is actually a cross correlation) Also Check out our premium deep learning course (in PyTorch) with end end to project: codebasics.io/courses/deep-learning-beginner-to-advanced
Just one doubt. How does it learn what will be the kernel matrix grid values. Coz the grid values of feature matrix were already provided by you, so how does it come to know by its own?
No University, No Online Certification courses, No other online platforms can match the simplicity and efficiency of Dhaval sir's teaching. Thank you very much for providing this for us.
Thanks Fousiya. Please check skillbasics.com, I have started adding quizs etc in some of the courses so that you can have guided learning. Check machine learning course on skillbasics.com (it is free)
Sir, I have seen your all playlists (Python, Machine Learning, Deep Learning, etc) and all are amazing. Your way of teaching and sharing knowledge proves how much you have expertise in data science. I recommend everyone to must go through all the videos if anybody wants to get an expert in data science.
I've gone through various online chapters and videos explaining the same concepts but none were as clear and easy to understand as this one. Kudos to you sir!
I Saw many videos to understand CNN Concept but non of them teaching from student point of view (student has no idea or basics), all are teaching little bit high level from student mindset, BUT REALLY AFTER I SAW UR VIDEO SIR, I Clearly understood without a doubt, THANKS A LOT & LOVE U. Keep it up making such more complex topics in simple way as u r doing now, really u r a boon for us once again love u sir.
simply amazing...I am doing PhD in DL and have come across several tutorials in youtube regarding CNN but none has been able to clarify my concept...you said the same things they said..but in much easier way so that everyone can understand..Now I can understand the meaning of every word i came across in my whole life about CNN....Hats off sir...PLS PLS PLS KEEP UP THE GOOD WORK
Sir same wording from my side. I doing PhD in image processing and learning CNN from different sources. Today's video I found complet, easy n conceptual. Thanku sir
I just wanted to say how impressed I am by the way the author of this video has explained such discrete and complex algorithms in such a simple manner, it is so easy to understand the concepts through the video, and all the examples that the author has put. Great work sir! I am really glad to have found such a channel where I can understand the basics of AI in such a simple and easy manner.
Thanks Arya for your feedback. My mission on this channel is exactly that (i.e. break down complex topics in a way that even a high school student can understand it easily), I will continue to upload videos like this with lots of visuals and easy explanation.
Hello sir iam your follower even though I'm taking course related machine learning I've about to start my carrier from data analyst please put some video how to create resume as a fresher what are the mistakes to avoid and how to build strong resume as a data analyst
I scoured the web for explanations of convolution, and found them to be instructions rather than explanations, or explanations put in terms that only people who already knew about the topic could easily understand. Your explanation was absolutely the most clear I've seen, and I grasped the concept halfway through. Congratulations! You should write a book and teach a course; you have a talent for explanation.
Steven, thanks for your kind words. I was able to leave my 9 to 6 job last year and now doing exactly as you said, i.e. building courses. Check codebasics.io
This is the best explanation of a CNN. I finally know what a stride means relative to my early thesis work in the 1990s using raw C/Unix. Now, with python, it seems like most tutorials just assume one knows a stride. But this really hits it good. Now, I know my 1990s kernel filters were done with a stride of 1. The pooling to reduce overfit is something I was completely unaware of despite two months of being in a course on medical imaging with CNN and deep learning. Your video is excellent for clarification.
Absolutely loved this intuitive explanation!! For me, getting into theory right at the beginning without an idea of what I'm trying to achieve in real life just bores and scares me. This explanation was the perfect summary of what to expect from CNNs. Thank you so much!
Brilliant teaching. This is what teachers all over country should strive to do in any subject. Here is a tough concept and you've broken it down to a level most educated people can digest. I find it (dare I say) even better than Andrew Ng's teaching style.
That You So Much From South Africa , Rural area, I followed the Entire Series on Your Deep Learning PlayList, Awesome Explanations. Unlike people who just jump into code or math formulas. Your Lectures truly give a person understanding. Much appreciated. God Bless you and your colleagues. Thank you.
You are incredible, simple, effective and straight to the point. Even the tune of your voice is like a music to ears, it shows a kind person with big heart and sincere attitude. With guys like you, what is the point of going to universities. Wish you all the best with great health and wealth and happiness
I watched many videos and read lot of stuff to understand CNN before watching this video. But this video is easiest and yet comprehensive way to understand CNN. Thank you for your efforts !!!
I'm very new to CNN, just started to learn this 4 days back. This video is the simplest one to grab the whole idea of what is CNN and how it actually works. Hats off to you, Sir! Keep 'em coming 🔥
I have been so behind in class trying to understand the complexity and architecture of CNNs and what is actually going on. This video made things amazingly clear and was simple to understand for such a complicated topic. Thanks!
How was I missing this channel ! So smooth presentation and interpretation. A big thanks to Dhaval sir and WQU putting out such videos directly from UA-cam.
I love you're explanation, very intuitive in how you go about things. as a testament to that fact, when you consider classifying an image of a monkey let's say - a monkey has an eye, ear, nose, mouth, hands, legs, and tail - you take the original image of your monkey and apply convolutional operation (filter operation) to create your convolutional map for these small parts that make up a monkey. the filter map is not restricted to just a 3 x 3 filter, you could have a 4 x 4 or a 5 x 5, etc. then from these individual filter maps you aggregate the maps to form a larger generalization for the monkey (does the part correspond to a body or a head perhaps?) to which you can infer the class of the animal to be a monkey or not. a very interesting video, and I'm sorry if my explanation doesn't do the concept justice but I'm pretty sure it's somewhere along those lines 🙂.
Dhaval Sir - I am a product of Indian education system - taught by teachers and professors in schools & colleges while was also trained by coaching institutes i.e. unorganized sector. I have also studied in the US for a short while. Hence, I can speak from experience. I would say you bring the best of all 3 worlds. Your style and delivery are so on point and effective that they are almost unmatchable! You have got this uncanny ability to connect the dots and create an effective transition from theory to the practical implementation of a concept. Please accept my heartiest thanks and gratitude for the immense service you're doing. You're hands down the best ML trainer at the moment on UA-cam. Hope to meet you personally sometime in India or the US. Your videos are awesome. Keep 'em coming !! :)
Arun, thanks for very detailed and useful feedback. This way I at least know my strengths and continue to produce more content using those. I myself struggled during my studies and even today whenever I try to understand complex subject (such as CNN), my whole goal is to make explanations extremely easy so that even a dumb person can understand. From your comment it looks like I have fulfilled part of that mission.
I've been confronting with a lot little intricate details which books were not clearly describing. I was so confused but this video has blown the all doubts. Outstanding way of explaining. Soft, simple and easy.
Dhaval, this is one of the most simplified explanations of CNN. I have watched many videos, but my concept got clear after watching this class. Thank you !!
Thanks a lot. Many online teachers considers as natural the conceptual operations that convolution does (edges, colors, etc) and just show a diagram with balls connected as if it explains everything 😜 You go deep enough by explaining clearly, and for that, thanks!
Superbly explained. I have recently started learning data science. Your video's are superb, Very well explained. I have finished all Machine learning videos. Started with Deep Learning. Most recommended channel for the beginners. Thank you soo much sir....... :)
![The moment we stopped understanding AI [AlexNet]](http://i.ytimg.com/vi/UZDiGooFs54/mqdefault.jpg)
Corrections and ammendments:
1) At 2:31, 6 mil * 4 mil will be a huge number in trillions. I put 24 million which clearly is a typo.
2) Convolution operation uses only the sum and not the average (that I have indicated in the video at 5:44. At 5:44 the correct value will be -1 and not -0.11)
3) In academia and popular literature the thing they refer as a convolution operation is actually a cross correlation. You can read more about it online but in AI world people call it convolution only so its ok to use that term. But just be aware that looking at it from purely mathematical standpoint, it is actually a cross correlation)
Also Check out our premium deep learning course (in PyTorch) with end end to project: codebasics.io/courses/deep-learning-beginner-to-advanced
Just one doubt. How does it learn what will be the kernel matrix grid values. Coz the grid values of feature matrix were already provided by you, so how does it come to know by its own?
No University, No Online Certification courses, No other online platforms can match the simplicity and efficiency of Dhaval sir's teaching. Thank you very much for providing this for us.
Thanks Fousiya. Please check skillbasics.com, I have started adding quizs etc in some of the courses so that you can have guided learning. Check machine learning course on skillbasics.com (it is free)
@@codebasics sir please provide a certificate with the courses. It will increase the interest & encouragement...
@@codebasics
Unable to understand image size?
True
He teaches very very clearly
Sir, I have seen your all playlists (Python, Machine Learning, Deep Learning, etc) and all are amazing. Your way of teaching and sharing knowledge proves how much you have expertise in data science. I recommend everyone to must go through all the videos if anybody wants to get an expert in data science.
Glad it was helpful! I appreciate you leaving a comment of appreciation
I can verify the fact that even a high school student can understand this topic. Well explained, I absolutely love your videos!!
ha ha .. thank you :)
Myth confirmed, a 59 year old can learn this too!
same here
I've gone through various online chapters and videos explaining the same concepts but none were as clear and easy to understand as this one. Kudos to you sir!
I Saw many videos to understand CNN Concept but non of them teaching from student point of view (student has no idea or basics), all are teaching little bit high level from student mindset, BUT REALLY AFTER I SAW UR VIDEO SIR, I Clearly understood without a doubt, THANKS A LOT & LOVE U. Keep it up making such more complex topics in simple way as u r doing now, really u r a boon for us once again love u sir.
One of the best videos on CNN! If only classroom teaching was as simple and easy as this, everyone would love to learn the complex concepts!
Thanks!
🙌🏾rabbits will help me remember pooling forever.
simply amazing...I am doing PhD in DL and have come across several tutorials in youtube regarding CNN but none has been able to clarify my concept...you said the same things they said..but in much easier way so that everyone can understand..Now I can understand the meaning of every word i came across in my whole life about CNN....Hats off sir...PLS PLS PLS KEEP UP THE GOOD WORK
Anindita thank you for your kind words of appreciation.
Sir same wording from my side. I doing PhD in image processing and learning CNN from different sources. Today's video I found complet, easy n conceptual.
Thanku sir
Good evening
I'm vinoth
I need some help for DL in welding process
I just wanted to say how impressed I am by the way the author of this video has explained such discrete and complex algorithms in such a simple manner, it is so easy to understand the concepts through the video, and all the examples that the author has put. Great work sir! I am really glad to have found such a channel where I can understand the basics of AI in such a simple and easy manner.
Thanks Arya for your feedback. My mission on this channel is exactly that (i.e. break down complex topics in a way that even a high school student can understand it easily), I will continue to upload videos like this with lots of visuals and easy explanation.
@@codebasics Looking forward to it then :P
I am a newbie exploring this field, hope to learn a lot from your channel!
Finally,You strarted CNN.
Really really thanks from bottom of heart.
Thank you so much sir.
Yup, many more exciting tutorials are coming up. Stay tuned
Hello sir iam your follower even though I'm taking course related machine learning I've about to start my carrier from data analyst please put some video how to create resume as a fresher what are the mistakes to avoid and how to build strong resume as a data analyst
Are you kidding me,,, you are an amazing lecturer. first time I have understood the layers I say !!
I scoured the web for explanations of convolution, and found them to be instructions rather than explanations, or explanations put in terms that only people who already knew about the topic could easily understand. Your explanation was absolutely the most clear I've seen, and I grasped the concept halfway through. Congratulations! You should write a book and teach a course; you have a talent for explanation.
Steven, thanks for your kind words. I was able to leave my 9 to 6 job last year and now doing exactly as you said, i.e. building courses. Check codebasics.io
No way, you can't make this topic so simple. You are a genius.
This is the best explanation of a CNN. I finally know what a stride means relative to my early thesis work in the 1990s using raw C/Unix. Now, with python, it seems like most tutorials just assume one knows a stride. But this really hits it good. Now, I know my 1990s kernel filters were done with a stride of 1. The pooling to reduce overfit is something I was completely unaware of despite two months of being in a course on medical imaging with CNN and deep learning. Your video is excellent for clarification.
Absolutely loved this intuitive explanation!! For me, getting into theory right at the beginning without an idea of what I'm trying to achieve in real life just bores and scares me. This explanation was the perfect summary of what to expect from CNNs. Thank you so much!
Brilliant teaching. This is what teachers all over country should strive to do in any subject. Here is a tough concept and you've broken it down to a level most educated people can digest. I find it (dare I say) even better than Andrew Ng's teaching style.
That You So Much From South Africa , Rural area, I followed the Entire Series on Your Deep Learning PlayList, Awesome Explanations. Unlike people who just jump into code or math formulas. Your Lectures truly give a person understanding. Much appreciated. God Bless you and your colleagues. Thank you.
You Cheated CNN was supposed to be hard!!! Thank you so much for such a wonderful video! Your explanation was so clear that I learned a lot today!
😅
This video was better than 3 weeks of lectures in my grad school course. A complete life saver.
This is very simple and amazing explaination for CNN. Full clarity on why CNN over ANN. Hats off to you !!
You are incredible, simple, effective and straight to the point. Even the tune of your voice is like a music to ears, it shows a kind person with big heart and sincere attitude. With guys like you, what is the point of going to universities. Wish you all the best with great health and wealth and happiness
Thanks for you kind words Saeed
Say no more, I have been watching a bunch of videos about CNN, so far this is the best. Just WOW! Thank you!
Honestly this is the best tutorial and explanation about the concept of CNNs!! absolutely loved it! Great job and thank you!!
I watched many videos and read lot of stuff to understand CNN before watching this video. But this video is easiest and yet comprehensive way to understand CNN. Thank you for your efforts !!!
Glad you liked it
I used to hate Machine Learning but you made me love it ...that is the power of a brilliant tutor!!
I'm very new to CNN, just started to learn this 4 days back. This video is the simplest one to grab the whole idea of what is CNN and how it actually works. Hats off to you, Sir! Keep 'em coming 🔥
☺️👍
I have been so behind in class trying to understand the complexity and architecture of CNNs and what is actually going on. This video made things amazingly clear and was simple to understand for such a complicated topic. Thanks!
How was I missing this channel ! So smooth presentation and interpretation. A big thanks to Dhaval sir and WQU putting out such videos directly from UA-cam.
I love you're explanation, very intuitive in how you go about things. as a testament to that fact, when you consider classifying an image of a monkey let's say - a monkey has an eye, ear, nose, mouth, hands, legs, and tail - you take the original image of your monkey and apply convolutional operation (filter operation) to create your convolutional map for these small parts that make up a monkey. the filter map is not restricted to just a 3 x 3 filter, you could have a 4 x 4 or a 5 x 5, etc. then from these individual filter maps you aggregate the maps to form a larger generalization for the monkey (does the part correspond to a body or a head perhaps?) to which you can infer the class of the animal to be a monkey or not. a very interesting video, and I'm sorry if my explanation doesn't do the concept justice but I'm pretty sure it's somewhere along those lines 🙂.
Thanks for the comment, you analogy is pretty good 😀👍🏼👍🏼
Dhaval Sir - I am a product of Indian education system - taught by teachers and professors in schools & colleges while was also trained by coaching institutes i.e. unorganized sector. I have also studied in the US for a short while. Hence, I can speak from experience.
I would say you bring the best of all 3 worlds. Your style and delivery are so on point and effective that they are almost unmatchable!
You have got this uncanny ability to connect the dots and create an effective transition from theory to the practical implementation of a concept.
Please accept my heartiest thanks and gratitude for the immense service you're doing. You're hands down the best ML trainer at the moment on UA-cam. Hope to meet you personally sometime in India or the US.
Your videos are awesome. Keep 'em coming !! :)
Arun, thanks for very detailed and useful feedback. This way I at least know my strengths and continue to produce more content using those. I myself struggled during my studies and even today whenever I try to understand complex subject (such as CNN), my whole goal is to make explanations extremely easy so that even a dumb person can understand. From your comment it looks like I have fulfilled part of that mission.
@@codebasics eagerly waiting for videos on RNN and recommendation engine
It is a best lecture on CNN in UA-cam.
I never seen a techer like u in youtube
You are my first youtube guru.
I have seen your machine learning series but i have to see again and again to make the topics store in my brain😋
👍👍
This is the best CNN explanation I have seen! Thanks for the video!
Just one line- The best CNN videos on UA-cam undoubtedly.
Great teaching Sir!
This gentleman's sense of humor is so pure. That even the most cringe jokes made me laugh. Good job Sir!
Well explained for beginners
great explanation for sure . ..
I've been confronting with a lot little intricate details which books were not clearly describing. I was so confused but this video has blown the all doubts.
Outstanding way of explaining. Soft, simple and easy.
This is the best video that explain CNN, even one paid course didn't explain it for me as you did.
best explanation of CNN . simple and straightforward
Excellent explanation of CNN!
Very nice video
Excellent skills of imparting knowlege
man i watched a lot of videos about CCN , i really got nothing, but after your video i am done with CNN, thank u so much man.
I have taken LinkedIn courses and few others and I must say, this is the best explanation I have found so far. Thank you Dhaval Sir 🙏
excellent class
You are awesome man.
Best resource on the UA-cam.
Love your work!!!!
Thanks.
Best explanation on CNNs ever!!!
I am actually a high school student(i am in 10th standard) and managed to understand it without any doubts, thank you so much sir!
😇👍👍
Excellent lecture, you have explained a complex concept in a very simple manner
Ye bunda Kamal Hai.
Iss terha kay log asal hero hotay hain jo career aur zindagi asan Banatay hain.
Dhaval, this is one of the most simplified explanations of CNN. I have watched many videos, but my concept got clear after watching this class. Thank you !!
beautifully explained. As a beginner i really got the idea of CNN just by watching this video. Thank you so much.
Beautiful Video. Good
Thanks a lot. Many online teachers considers as natural the conceptual operations that convolution does (edges, colors, etc) and just show a diagram with balls connected as if it explains everything 😜 You go deep enough by explaining clearly, and for that, thanks!
BEST EXPLANATION OF CNN EVER!!!
Amazing! High Quality Content.
Excellent teaching Mr. Dhaval. Thoroughly impressed the way you make things simple in teaching!!! Hats off to you sir!!!!
🤗👍👍👍🙏
Nice explanation 👌 👍 👏 😀
Myth confirmed: Even a high schooler can learn this.
☺️☺️☺️👍
Amazing Video. Great Explanation.
our college lecturers need to learn from this man how to teach concepts
You people are asset to this country.......very well explained.....👍
As always very simple and clear.
Superbly explained. I have recently started learning data science. Your video's are superb, Very well explained. I have finished all Machine learning videos. Started with Deep Learning. Most recommended channel for the beginners. Thank you soo much sir....... :)
Mind blowing explanation🤯🤯
Good job amazing explanition
Glad it was helpful!
best video for explanation.
I like your teaching way sir.
Simply superb 👌👌
This is the best neural network video I have seen. You are the best
AMAZING....such a quality content
Going to watch all videos
One of THE best explanations I have come across . Thank you !!
One of the best videos for CNN
Love your explanation almost without complicated math, or even without math despite so high level topic as CNN 😁 it’s talent 👏
Glad you liked it!
very good explaination
Amazing Content 👏👏
Nice explanation👍👍
extraordinary explanation. understand very clearly. best class for cnn
Glad you liked it
You're videos really helped me in final year research mate! Thank you very much!
Great Explanation and wonderful.
Good introduction to the topic
Glad it was helpful!
Just brilliant.. 👏👏👏
Excellent explanation of CNN.
👍😊
Perfectly explained 🔥
THIS IS SOOOO GOOOODDD. SUCH A NICE EXPLANATION
Best cnn tutorial..
Excellent. Watched again!!
This is amazing thank you! Simply explained, my professor failed to explain and a lot other youtuber failed as well.
Extremely clear and easy to understand. Thank you very much
Your tutorials are gold! Thank you so much for your dedication
Glad you like them!
Great Sir...it is life svaing ...in exam time
Such high quality content
This is what I am looking for! Need to know first the fundamentals. Thank you very much sir!
Thank You Sir... Amazing Explanation Scala.....
Very well explained👍
Best teacher ever to be on this planet 🌎