LSTM Recurrent Neural Network (RNN) | Explained in Detail

Поделиться
HTML-код
  • Опубликовано: 7 фев 2025

Комментарии • 99

  • @hafolahbi
    @hafolahbi 2 месяца назад +4

    I can't pass without commenting and liking this video. It is invaluable, far more than reading it in journals

    • @MachineLearningWithJay
      @MachineLearningWithJay  Месяц назад +2

      Wow.. thank so much for such an amazing comment! Means a lot to me. Glad I could help. I wish you all the very best for your exams!

  • @srishti6637
    @srishti6637 Год назад +8

    best explanation with no faltu pnchyti and made the topic crystal clear

  • @ivana_ftn
    @ivana_ftn Год назад +33

    You are better than my professor, thank you

  • @dmg-s
    @dmg-s 2 года назад +11

    I am so happy now. Thanks!

  • @eng.mohamedemam6489
    @eng.mohamedemam6489 Год назад +4

    from a man from Egypt send big thanks to you ❤❤

  • @ikraamhanif7966
    @ikraamhanif7966 6 месяцев назад +2

    Kash mene phly prh liya hota aj paper hai aur apki videos dekh kr itna achay sy smjh arha hai kiya hi bolun apko...Thank u so much for providing us knowledge like this

  • @Ayocogn
    @Ayocogn 5 месяцев назад +2

    I think u deserve much more than 27k subscribers man. I totally got it after watching this playlist

    • @MachineLearningWithJay
      @MachineLearningWithJay  4 месяца назад

      Hehe! Thank you so much! I appreciate it, and I agree. Hopefully with more videos and your support, the channel might grow.

  • @rajkamal1705
    @rajkamal1705 Год назад +3

    Very easy to understand. You are better than many prof. Thanks bro.

  • @jayhu6075
    @jayhu6075 Год назад +9

    I am very glad to find your channel. You make this topic for a beginner as me so understandable.
    Hopefully a following to write this in a python code. Many thanks.

  • @tridibeshmisra9426
    @tridibeshmisra9426 Год назад +2

    Best explanation...... It helped me for endsem exam...thank u sir.....keep creating ... let's get riding🙂🙂

  • @s8x.
    @s8x. 9 месяцев назад +2

    brother u made learning machine learning so easy. When i got money i will be sure to show my thanks

    • @MachineLearningWithJay
      @MachineLearningWithJay  3 месяца назад

      @@s8x. haha… you appreciating this is enough for me me! Goad I could help!

  • @nikhils2155
    @nikhils2155 5 месяцев назад +5

    Thank you for these awesome classes brother

  • @christianfaust5141
    @christianfaust5141 4 месяца назад +1

    Thank you from Germany, I appreciate your work

  • @deepurachakonda28
    @deepurachakonda28 2 года назад +4

    u made my day..thnx lot

  • @rezamohammadi1140
    @rezamohammadi1140 Год назад +1

    Very helpful because of mathematical explanation and summery in the last

  • @MurodilDosmatov
    @MurodilDosmatov 6 месяцев назад +1

    Thousand of thanks for your effort to make this video tutorial

  • @pavangoyal6840
    @pavangoyal6840 Год назад +2

    Excellent. Thank for this video and explaining complex concept like LSTM in very short and crisp video

  • @khushiyadav-st4oh
    @khushiyadav-st4oh 3 месяца назад +1

    Best Explanation Ever

  • @Thing1Thing11
    @Thing1Thing11 Год назад +2

    Thank you so much! I was so lost and you really helped me get to grips with what is going on

  • @arvinflores5316
    @arvinflores5316 3 года назад +2

    That was a good binge man. Hopefully attention/transformers will be covered too!

  • @dinushachathuranga7657
    @dinushachathuranga7657 2 месяца назад +1

    Thanks for the clear explanation ❤❤

  • @Bunches_of_Entertaiment
    @Bunches_of_Entertaiment 2 года назад +2

    Superb explanation brother.. thank you so much 😍.. I got very clear understand on LSTM and as well as RNN

  • @muhammadumarwaseem
    @muhammadumarwaseem 11 месяцев назад

    Thank you for the detailed to the point explanation.

  • @kavoshgar9733
    @kavoshgar9733 11 месяцев назад

    You describe everything very well✌️

  • @NeroFernando
    @NeroFernando 20 дней назад

    Hi Jay, thank you for the great explanation! I noticed at 18:30 that 𝑜𝑡 is used for both the output gate and the softmax-activated output of the cell. Wouldn’t it be clearer to use a different name, like 𝑦𝑡, for the cell’s final output to avoid confusion? Thanks again for the amazing content!

  • @vinayakmane7569
    @vinayakmane7569 Год назад +4

    remarkable explanation , keep bringing good content. just one little suggestion , try to write keypoints on board while explaining so that we can copy and it will help us while revising

    • @MachineLearningWithJay
      @MachineLearningWithJay  3 месяца назад

      @@vinayakmane7569 Write may be I will try to keep the summary at the end and write key points there? Would it help? Anyways I will keep this in mind for my future videos

  • @madhusushma968
    @madhusushma968 25 дней назад +1

    sir plz dont use black background and red pen its realy hard to see try another combination or increase the thickness of the pen
    and your explanation as always 20/10 (good job)

    • @MachineLearningWithJay
      @MachineLearningWithJay  25 дней назад

      @@madhusushma968 Hi, thank you for the suggestion. I will keep this in mind

  • @saatvikkool
    @saatvikkool 3 года назад +2

    You're doing a great job bro ✌️❤️

  • @hamedhasani5774
    @hamedhasani5774 3 месяца назад

    awesome work

  • @KulkarniPrashant
    @KulkarniPrashant 9 месяцев назад

    Clear and concise explanation, thank you!

  • @INCREDIBLE9463
    @INCREDIBLE9463 7 месяцев назад

    Very good explanation

  • @harshitraj8409
    @harshitraj8409 2 года назад +2

    Great Explanation

  • @sandeepkomalpothu44
    @sandeepkomalpothu44 3 года назад +2

    thanks man …. very helpful … cheers !!!

    • @MachineLearningWithJay
      @MachineLearningWithJay  3 года назад +1

      Cheers!!

    • @sandeepkomalpothu44
      @sandeepkomalpothu44 2 года назад

      @@MachineLearningWithJay could you please make videos on GRUs ,se2seq,Attention models, Transformers?

    • @MachineLearningWithJay
      @MachineLearningWithJay  2 года назад +1

      @@sandeepkomalpothu44 I am goong to upload those videos, but it will take some time. Right now, i am busy with my exams, so will upload when i get time. 😇

  • @juditmaymo9714
    @juditmaymo9714 Год назад +2

    love this video!!

  • @QuratRaja-q2v
    @QuratRaja-q2v Год назад

    Worthy explanation!

  • @evildead3734
    @evildead3734 3 года назад +1

    Great resources 🙌

  • @PRAJAKTAPATKAR-t7z
    @PRAJAKTAPATKAR-t7z 10 месяцев назад

    Amazing- great work

  • @this_is_the_baat
    @this_is_the_baat 7 месяцев назад

    thanks my dear bro

  • @ABCD-wd1sk
    @ABCD-wd1sk 2 месяца назад +1

    Thanks

  • @Animelover-oo7cz
    @Animelover-oo7cz 9 месяцев назад

    you are the best thank youu

  • @7aanusha885
    @7aanusha885 Год назад

    thank you for this video

  • @user_userovich
    @user_userovich Год назад

    Thank you very much!

  • @shaveta01
    @shaveta01 10 месяцев назад

    Good vdo
    Pl use white marker 😊instead of red

  • @achreflegend4335
    @achreflegend4335 6 месяцев назад

    5:00 the green formula ; shouldn't it be a sum of multiplication through 0~t instead of just a sum

  • @iftikharullah3616
    @iftikharullah3616 Год назад

    I watched many videos for this topic but couldn't understand it. You made every point clear in a beautiful way 🫡

  • @slingshot7602
    @slingshot7602 11 месяцев назад

    Tnx a lot

  • @VenomOmpi
    @VenomOmpi 9 месяцев назад

    thanks a lot!

  • @vishnusit1
    @vishnusit1 Год назад

    At 8:45 you are wrong about matrix multiplication

  • @KishorA-s8f
    @KishorA-s8f Год назад

    Thank you

  • @mainakmukherjee3444
    @mainakmukherjee3444 Год назад

    Hello bhaiya, thanks for the informative contents. But can you please explain me why you are saying ft (forget gate) is a matrix. From the formula, it just an output of a sigmoid function, which I think a scalar value for each time step. Please explain this part. 🙏🙏

  • @pavangoyal6840
    @pavangoyal6840 Год назад

    For time series data (there are around 60 input variables) and there are two outputs variables. Which deep learning model would be best LSTM ? Here accuracy matters. For learning time does not matter. For 2 output variables how to design LSTM model ?

  • @rohit1109p
    @rohit1109p 9 месяцев назад +1

    What is Bc, Bf, Bi, Bo added everywhere ?? Is that bias ?

    • @forsakennawfal546
      @forsakennawfal546 7 месяцев назад

      Biases or as we can say the Constants added in every function, indicating the margin of error.

  • @KhadijaTulKubra-o2o
    @KhadijaTulKubra-o2o Год назад

    Hi! i want to learn text detection from images using RNN. Please if you can help ???

  • @charanm1773
    @charanm1773 Год назад

    Arigatto

  • @tanvirtanvir6435
    @tanvirtanvir6435 Год назад

    3:22 problem with RNN

  • @beshosamir8978
    @beshosamir8978 2 года назад +1

    Hi , i hope one day you explain transformers because ur explainations is great ,but i really need your help about something because i really got tired of searching about the answer
    i got stucked on something and seems like no one helps so i hope you help me , now i studied about LSTM and Bi-LSTM and i understood them well , but i read some blogs said that bi directional LSTM good for sentiment analysis and time series so i really got confused about it , How it could be useful !!!! it will be useful if my current prediction depends on what happens in the future ,so how it could be usefull in sentiment analysis if i already will predict my final output in the last word so there is no future because i stand in the last , i know it could be usefull in some applications like name entity recognation because the type of the output is (many) so maybe my current output depends on what is happend in the future
    i really hope to help me because i didn't find any reason after 2 hours of searching in google

    • @MachineLearningWithJay
      @MachineLearningWithJay  2 года назад +1

      Hi, bidirectional LSTM looks at all the words appearing in an input sentence, from both directions, front to back and back to front. So, you can always assume that bidirectional LSTM can be useful for any application involving sentences as input. I don’t have any more reason for this, for now.

    • @beshosamir8978
      @beshosamir8978 2 года назад

      @@MachineLearningWithJay
      So , is my question make sense ? or there's something i can not understood about the intuation for now ?

  • @tahsinulkarim
    @tahsinulkarim Год назад

    ❤❤❤

  • @martinant44
    @martinant44 9 месяцев назад

    te quiero mucho

  • @aitoquantum
    @aitoquantum Год назад

    the summation formula is wrong! one product sum should be there along with plain summation, that will unfold the longer terms...

  • @slingshot7602
    @slingshot7602 11 месяцев назад

    GRU???

  • @MrKhanRizwan
    @MrKhanRizwan 4 месяца назад +1

    Too complex must be targeted for mathematicians

    • @MachineLearningWithJay
      @MachineLearningWithJay  4 месяца назад

      Hi, yeah… the details LSTM is a bit complicated. It doesn’t make much sense at first, and you need mathematical background to understand this.
      If you are not concerned with the implementation details, then you can directly use LSTM through frameworks like PyTorch, Keras

  • @shantanusingh2198
    @shantanusingh2198 6 месяцев назад +1

    Proably the worst video on lstm i have seen

    • @MachineLearningWithJay
      @MachineLearningWithJay  6 месяцев назад

      Amm.... sure... All videos are different kind... sad to see this tutorial wasn't help to you :(