Groups, Depthwise, and Depthwise-Separable Convolution (Neural Networks)

Поділитися
Вставка
  • Опубліковано 10 гру 2024

КОМЕНТАРІ • 93

  • @sajanphilip8221
    @sajanphilip8221 Рік тому +78

    Please don't stop making videos. They are of great help. Thank you for your efforts.

  • @slime67
    @slime67 Рік тому +6

    amazing! as an AI researcher I missed these videos back in the days when I studied convolutions, hope they'll bring more understanding to the people just coming to the field!

  • @msergejev
    @msergejev 2 місяці тому +1

    An absolute pinnacle of online education materials in the field, when it comes to giving a real gut intution of what do operation look like 🙌 its a real talent you got there, thank you on behalf of the rest of the internet for using it well

  • @alirezamohammadi-j7f
    @alirezamohammadi-j7f 8 місяців тому

    I just wanted to say a huge THANK YOU for all the incredible animations you've been creating. Your work has been a game-changer for me, making complex concepts so much easier to understand.

  • @crazykidsstuff
    @crazykidsstuff 9 місяців тому +1

    This is amazing! There is a lot of great material out there, and your channel is a really solid and valuable contribution to that. Thanks a batch!

  • @NoMusiciansInMusicAnymore
    @NoMusiciansInMusicAnymore 2 місяці тому

    This helped so much, you can't understand how thankful I am

  • @SajjadZangiabadi
    @SajjadZangiabadi Рік тому +1

    Absolutely loved the way the instructor used animations to explain concepts like Groups, Depthwise, and Depthwise-Separable Convolution. It made understanding the topic so much easier and engaging. Keep up the great work!

  • @hashimsharif1542
    @hashimsharif1542 10 місяців тому

    By far the best explanation of depthwise-separable convolutions I found! This is a service, thanks!

  • @mynameisvahe
    @mynameisvahe Рік тому +1

    One of the best channels on DL I’ve seen so far. Please publish more!

  • @danirubio2847
    @danirubio2847 Рік тому +1

    This was an incredible video. You can see the the amount of work and dedication; and you explain really good! Thanks, please keep on doing this videos

  • @lakshaysethi7445
    @lakshaysethi7445 Рік тому

    Great Work. I am a Master's student in ML and I your animations are really helpful in understanding this concept!! Thanks a lot.

  • @j________k
    @j________k Рік тому

    Great stuff... the algorithm should give your content more attention!

  • @linkernick5379
    @linkernick5379 Рік тому

    This is so underestimated channel, will share it as much as I can. Thank you, Mr AI Animator!

  • @yousrakateb2383
    @yousrakateb2383 6 місяців тому +2

    Please continue make such amzing videos....they really helped me

  • @CBhattacharjee
    @CBhattacharjee Рік тому

    This is great stuff. Please continue to make more, you are saving new scholar lives here!

  • @homakashefiamiri3749
    @homakashefiamiri3749 Рік тому

    The best conv layer visualization so far👍
    Thank you for your great work💥💥

  • @Mars-xm7uz
    @Mars-xm7uz 8 місяців тому

    Your work is truly amazing, please keep enlighting us!

  • @Scarlety8109
    @Scarlety8109 2 місяці тому

    game changer for anyone learning neural networks

  • @ShadeAKAhayate
    @ShadeAKAhayate Рік тому

    Thanks, the explanation of this mechanic is exceedingly lucid.

  • @katuz9366
    @katuz9366 Рік тому

    Hey, I want to thank you for spending time making great content animated! I’ve been using depth wise for a time and it has been always a little hazy

  • @josephpark2093
    @josephpark2093 Рік тому

    You are a freaking saint. I gotta sub for the effort you put in.

  • @janphiliprichter96
    @janphiliprichter96 8 місяців тому

    Incredible video! Brilliant visualisations and perfect explanation. Keep it up

  • @leonardommarques
    @leonardommarques 6 місяців тому

    You got a new subscriber. You are 3b1b of AI. Thanks for existing.

  • @kozhushko
    @kozhushko День тому

    Thank you! Such a great video!

  • @rafa_br34
    @rafa_br34 5 місяців тому

    Incridibly helpful, keep up the good work!

  • @harrydawitch
    @harrydawitch 3 місяці тому

    My second favourites 3Brown1Blue channel

  • @pi5549
    @pi5549 Рік тому

    Do remember in future vids to invite viewers to smash the like button, as it improves your ranking as per the Algorithm. I just realised I watched half a dozen of these without hitting it.

  • @yanickpalmers
    @yanickpalmers Рік тому +1

    absolutely love these videos! doing gods work

  • @alexandrostsagkaropoulos
    @alexandrostsagkaropoulos Рік тому

    That was so intuitive. Thanks for that!

  • @nini_kys
    @nini_kys 11 місяців тому

    Amazing video, so well explained and to the point.

  • @mingwen3854
    @mingwen3854 10 місяців тому

    Very intuitive to understand, thank you.

  • @himanshu1373
    @himanshu1373 Рік тому

    thankyou very much brother this video means a lot for people like me 😍

  • @sk7w4tch3r
    @sk7w4tch3r 10 місяців тому

    Thanks!

  • @KamleshKumar-s9n3g
    @KamleshKumar-s9n3g Рік тому

    Thank you for making videos like this.

  • @sharjeel_mazhar
    @sharjeel_mazhar Рік тому

    Please sir, also make visualizations like these on RNN, LSTM and most importantly Transformers. Would be really thankful to you. And also, your videos on CNN are just gems in the ocean of youtube.

  • @matheuscardosomoraes5641
    @matheuscardosomoraes5641 Рік тому

    Congratulations for amazing class

  • @revimfadli4666
    @revimfadli4666 Рік тому +2

    Fantastic explanations, even though I understand the paper diagrams, this makes it super clearer. Would you cover cascaded/DenseNet someday?

  • @이정민-n6g
    @이정민-n6g Рік тому

    this animation really helps me , thanks!

  • @petyap7600
    @petyap7600 Рік тому

    Finally understood. Thanks. Really helpful videos.

  • @guyindisguise
    @guyindisguise Рік тому

    Another great video. Can't wait for you to go into animating Transformers!

  • @deeps-n5y
    @deeps-n5y Рік тому

    underrated channel

  • @happyTonakai
    @happyTonakai 6 місяців тому

    This is so great!

  • @甩册
    @甩册 Рік тому

    Thank you very much for your sharing. It helps me a lot. I would really appreciate it if you could add subtitles

  • @troyhan3741
    @troyhan3741 Рік тому +1

    I want to know how can you make this video , what tools of software you used ?

  • @sinkseeker
    @sinkseeker Рік тому

    Hi Animated AI, for clarification, are the stacks of cubes in the first 30 seconds of the video feature maps? Also, how exactly did the depth increase as we get into the deeper layers? Based on my understanding, the lecture you provided was focused more on maintaining the depth while increasing its efficiency. I hope to hear from you soon! Your work is great!

    • @animatedai
      @animatedai  Рік тому

      That's correct, they're the feature maps which are the inputs/outputs of the layers.
      The depth of a feature map is equal to the number of filters in the convolutional layer that created it. So the depth increases that you're seeing are simply layers that have more filters than the number of features in their input. Let me know if that isn't what you meant by your question.
      This video shows the depth staying the same in a depth-wise separable convolution, but you can still depth-wise separate a layer that increases the number of filters and get the performance benefits. You can just take the input depth and use twice (or some other multiple of) that value for the filter count in both the depth-wise and point-wise convolutions.

    • @sinkseeker
      @sinkseeker Рік тому

      @@animatedai I see, I see. So if the input is an RGB image, and the first convolutional layer uses 5 filters, then the depth of the feature map will be 5. If that feature map goes to another convolutional layer with 5 filters, will the output contain a feature map with a depth of 25?

    • @animatedai
      @animatedai  Рік тому

      In that example, both outputs would have a depth of 5 because both layers have a filter count of 5. My video on filter count might help you visualize the relationship there: ua-cam.com/video/YSNLMNnlNw8/v-deo.html
      These videos are both part of this playlist on convolution: ua-cam.com/play/PLZDCDMGmelH-pHt-Ij0nImVrOmj8DYKbB.html

    • @sinkseeker
      @sinkseeker Рік тому

      @@animatedai Hi animatedAI! I'll check the link out. I hope I'll get it afterwards haha. Thanks for sharing!

  • @abdulrahmankhalaf2982
    @abdulrahmankhalaf2982 11 місяців тому

    Awesome job, I have a quesion out of the box, how you are did this work? which programs used in this video to produce it?

  • @naasvanrooyen2894
    @naasvanrooyen2894 Рік тому

    Thanks! Best explanation ever

  • @AvesNova
    @AvesNova Рік тому

    Great video! Keep up the good work

  • @ayeletgarber3697
    @ayeletgarber3697 9 місяців тому

    תודה!

  • @DashankaWork
    @DashankaWork 7 місяців тому

    Great video... keep it goining..Thanks a lot

  • @alexgamingtv7118
    @alexgamingtv7118 Рік тому

    Truly awesome!

  • @-mwolf
    @-mwolf Рік тому +1

    Thanks! Could you also cover convolutions with processing audio?

  • @valovanonym
    @valovanonym Рік тому

    Great vidoe! Your website will be a very usefull ressource.
    May I ask you what tool you are using for creating these animations?

    • @animatedai
      @animatedai  Рік тому

      I'm using Blender and relying heavily on the Geometry Nodes feature.

  • @thivuxhale
    @thivuxhale Рік тому

    this is too cool to handle!!!!

  • @adrienkin
    @adrienkin 8 місяців тому

    From your example, it could be nice to give the number of computations as example of +/- 9x faster :)

  • @joaomartinho4845
    @joaomartinho4845 Рік тому +1

    so so good

  • @panjak323
    @panjak323 Рік тому

    Soo the output has the same number of channels as the input? Or can you modify that by 1x1 convolution at the end ? Also doesn't this double the required storage for feature maps ?

    • @animatedai
      @animatedai  Рік тому

      In practice, it probably doesn't make a huge difference where you increase the number of channels. You could increase the channels in the depthwise convolution as long as you wanted the output channel count to be a multiple of the input. EfficientNet actually increases the number of channels with an extra pointwise convolution before the depthwise convolution.
      Yes, it increases the storage required during training in TensorFlow and PyTorch. Post-training, you don't necessarily need to keep around all the intermediate feature maps. So whether or not it doubles the required storage is dependent on the library (if any) that you're using for deployment.

  • @pa7235
    @pa7235 Рік тому

    great material!

  • @xXMaDGaMeR
    @xXMaDGaMeR Рік тому

    Great work, thank you

  • @dimitrisspiridonidis3284
    @dimitrisspiridonidis3284 Рік тому

    amazing job !!!

  • @commanderlake7997
    @commanderlake7997 7 місяців тому

    It should be noted that this will not scale well with tensor cores and may even be slower.

  • @mathieugalle391
    @mathieugalle391 Рік тому

    Great work, thanks !
    Maybe FPNs next time ? :-D

  • @raiswea9319
    @raiswea9319 23 години тому

    What are groups?

  • @AnonymousIguana
    @AnonymousIguana Рік тому

    Fantastic!

  • @darmstadtbeste4590
    @darmstadtbeste4590 Рік тому

    awesome thanks!

  • @dennervasconcelosrodrigues1212

    Thank you

  • @DanielTorres-gd2uf
    @DanielTorres-gd2uf Рік тому

    Nice job!

  • @martinhladis1941
    @martinhladis1941 6 місяців тому

    Excelent!!

  • @dandoe2461
    @dandoe2461 Рік тому

    great video

  • @grayboywilliams
    @grayboywilliams Рік тому

    Thank you!

  • @tomoki-v6o
    @tomoki-v6o Рік тому

    Can we convert trained standard Convnets to depth wise ones ?

    • @animatedai
      @animatedai  Рік тому

      You could theoretically separate any kernel into a depthwise-separated one. But you'd need a lot more filters in the depthwise convolution, so the result would be about the same performance. The performance improvement comes from training the network to take advantage of depthwise-separated convolutions.

  • @martinm8179
    @martinm8179 Рік тому

    These videos are excellent, but I suspect your ability to discern adjacent colors on a color wheel greatly outpaces mine. I have to pause and stare back and forth between blocks. It would be nice it were easier to see. Tools like Viz Palette can help pick better colors for data visualization.

    • @animatedai
      @animatedai  Рік тому +1

      I appreciate your feedback! I could rant for hours about how hard it is to pick colors :). I have two clarification questions: 1) Which part of the video did you pause to stare back and forth between blocks? 2) Which feature of Viz Palatte do you think would have helped pick colors for that instance?

  • @mayapony
    @mayapony 4 місяці тому

    thx!!

  • @clutchplayz1180
    @clutchplayz1180 22 дні тому

    fye

  • @xXMaDGaMeR
    @xXMaDGaMeR Рік тому

    ok visualize transforms next please, Vision Transforms would be nice.

  • @cardinalblues7121
    @cardinalblues7121 Рік тому

    please be more productive . Your videos are amazing

  • @a.h.s.2876
    @a.h.s.2876 10 місяців тому

  • @ucngominh3354
    @ucngominh3354 Рік тому

    hi

  • @alexvillalobos8245
    @alexvillalobos8245 5 місяців тому

    jiff

  • @xXMaDGaMeR
    @xXMaDGaMeR Рік тому +7

    ok visualize transforms next please, Vision Transforms would be nice.