How to implement Linear Regression from scratch with Python

Поділитися
Вставка
  • Опубліковано 7 лис 2024

КОМЕНТАРІ • 82

  • @Artificial_Intelligence_AI
    @Artificial_Intelligence_AI Рік тому +11

    This channel is amazing ❤. This is the type of content a lot of instructors forget to teach you when you’re learning ML but this girl explains everything very well from scratch. Congratulations for your content and I hope to watch more of your videos, you deserve more views for this incredible job.

  • @afizs
    @afizs 2 роки тому +7

    I have used Linear Regression many times, but never implemented from scratch. Thanks for an awesome video. Waiting for the next one.

    • @AssemblyAI
      @AssemblyAI  2 роки тому +1

      That's great to hear Afiz!

  • @tienshinhan8189
    @tienshinhan8189 10 місяців тому +1

    I rarely bother commenting because I'm rarely impressed. This video was amazing. I love that you're showing theory and OOP. Usually I see basic definitions and code all in one script.

  • @MU_2000
    @MU_2000 5 місяців тому +2

    Thank you for this video. Before i wached it i spend a couple of day to understand how to make custome code without frameworks ))

  • @bendirval3612
    @bendirval3612 2 роки тому +7

    Wow, that really was from scratch. And the hardest way possible. But it's perfect for teaching python. Thanks!

  • @lamluuuc9384
    @lamluuuc9384 Рік тому +13

    I think there is a problem of missing the 2 multiplication in calculating dw, db in the .fit() method:
    dw = (1/n_samples) * np.dot(X.T, (y_pred-y)) * 2
    db = (1/n_samples) * np.sum(y_pred-y) * 2
    If we follow the slides, it's not absolutely wrong but it can affect the learning rate

    • @priyanshumsharma4276
      @priyanshumsharma4276 Рік тому

      No, after derivation, there is no square.

    • @zh7866
      @zh7866 Рік тому

      The comment suggests multiplication with 2 based on derivation rule NOT square

    • @jhicks2306
      @jhicks2306 9 місяців тому

      Agree with this! Also, thanks, Misra, for the awesome video series.

    • @surajsamal4161
      @surajsamal4161 7 місяців тому +2

      @@priyanshumsharma4276 he's not talking about the square he's talking about the square which came down cause of derivative

  • @haseebmuhammad5145
    @haseebmuhammad5145 2 місяці тому

    I was seeking such kind of lectures. Lectures are awesome

  • @l4dybu9
    @l4dybu9 2 роки тому +6

    Thank u so much for this video. 💖💖
    It's makes us feel more confident when we know how to do it drom scratch than using libraries ✨

  • @markkirby2543
    @markkirby2543 Рік тому +2

    This is amazing. Thank you so much for all your clear explanations. You really know you stuff, and you make learning this complex material fun and exciting.

  • @ayo4590
    @ayo4590 Рік тому +2

    Thank you so much, this video really helped me get started with understanding machine learning algorithms. I would love if you could do a video on how you would modify the algorithm for multivariate linear regression.

  • @KelvinMafurendi
    @KelvinMafurendi 3 місяці тому

    Great explanations indeed. The transition from theory to implementation in Python was awesome!
    Say, is this a good starting point for a beginner in Data Science or I should stick to the out-of-the-box sklearn methods for now?

  • @GeorgeZoto
    @GeorgeZoto Рік тому +6

    Another amazing video! Slight typo in the definition of matrix multiplication and =dw part as well as an omission on the constant 2 (which does not effect calculations much) in the code when you define the gradients but other than that this is beautiful 😃

  • @kamranzamanni6826
    @kamranzamanni6826 Місяць тому

    Thanks for making it easy to follow along!

  • @AndroidoAnd
    @AndroidoAnd 8 місяців тому +1

    I am not sure if you copied this code from Patrick Loeber. He has a youtube video with the same code posted years ago. If you did, please give credit to Patrick.
    This is the name of his video: Linear Regression in Python - Machine Learning From Scratch 02 - Python Tutorial

  • @compilation_exe3821
    @compilation_exe3821 2 роки тому +3

    YOU guys are AWWWWESOME

  • @1000marcelo1000
    @1000marcelo1000 2 роки тому +1

    Amazing video! I learned so much from it! Congrats!!!
    Could explain more detailed all of this and show next steps like "where" and "how" this can be implemented further in some scenarios?

  • @harry-ce7ub
    @harry-ce7ub 8 місяців тому +2

    Do you not need to np.sum() the result of np.dot(x, (y_pred-y)) in dw as well as multiply by 2?

  • @luis96xd
    @luis96xd 2 роки тому +3

    Amazing video, I liked the code and the explanations, it was easy to read and understand, thanks! 😁👍👏💯

  • @1Rashiid
    @1Rashiid Місяць тому

    perfect explanation! Thank you.

  • @KumR
    @KumR 9 місяців тому

    Wow... Awesome.. Do you have something like this for Deep learning algo ?

  • @sajinsanthosh9536
    @sajinsanthosh9536 2 роки тому +2

    followed the tutorial exactly right, but still different. Using trial version. Thank you*

  • @srinivasanshanmugam9527
    @srinivasanshanmugam9527 10 місяців тому

    Thanks for this video, May I know , why we need to run in a for loop for 1000 iterations?

  • @سعیداحمدی-ل1ث
    @سعیداحمدی-ل1ث Рік тому

    ادامه بده دختر کارت عالیه ممنون ازت

  • @mohamedsadeksnoussi8479
    @mohamedsadeksnoussi8479 2 роки тому +4

    The content is really insightful, but for dw and db, should it be (2/n_samples) instead of (1/n_samples) ?

    • @adarshtiwari7395
      @adarshtiwari7395 2 роки тому +2

      did you figure out why she used the equations without 2 in the video?

    • @mohamedsadeksnoussi8479
      @mohamedsadeksnoussi8479 2 роки тому +4

      Hi@@adarshtiwari7395, unfortunately no. But (2/n_samples) appears to be correct.
      I checked out other resources and all of them used (2/n_samples).
      You can even try it by yourselves, (1/n_samples) doesn't affect the model behavior (performance), but from my point of view, it's incorrect.

    • @prashantkumar7390
      @prashantkumar7390 Рік тому +3

      @@adarshtiwari7395 It doesn't change our actual optimisation problem, I mean when the mean_square_loss is minimised, any scalar*mean_square_loss will be minimised. Hence using 2 or not doesn't make a difference at all.

    • @mohammedabdulhafeezkhan4633
      @mohammedabdulhafeezkhan4633 Рік тому +2

      @@prashantkumar7390 But it is better to use 2 in the equation, even though it's a constant and doesn't affect the outcome in a major way.

    • @prashantkumar7390
      @prashantkumar7390 Рік тому +1

      @@mohammedabdulhafeezkhan4633 it doesn't affect the outcome AT ALL.

  • @MahmouudTolba
    @MahmouudTolba 2 роки тому +1

    your code is clean

  • @AbuAl7sn1
    @AbuAl7sn1 4 місяці тому

    thanks lady .. that was easy

  • @m.bouanane4455
    @m.bouanane4455 9 місяців тому +2

    The gradient calculation lacks the multiplication by coefficient 2, I guess.

  • @LouisDuran
    @LouisDuran 7 місяців тому

    WHat check could be added in the case that the model is not converging to the best fit within the n_iters value?

  • @kreativeworld2073
    @kreativeworld2073 2 роки тому

    Thank you so much.
    It really helped me understand the entire concept:)

  • @parthaguha4705
    @parthaguha4705 Рік тому +1

    Awesome explanation😇

  • @okefejoseph6825
    @okefejoseph6825 5 місяців тому

    Is there a way I can get the slide?

  • @rajesh_ramesh
    @rajesh_ramesh 3 місяці тому

    the gradient part actually misses the coefficient 2 in both differentiations (dw, db) and y - y^

  • @rbodavan8771
    @rbodavan8771 2 роки тому +2

    how you designed your video templets is awesome.
    but it would be good for learner if you also post it on kaggle notebook and link it each other. sometimes it's happened to me like best to read then watch. let me know your thought

    • @AssemblyAI
      @AssemblyAI  2 роки тому +1

      You can find the code in our github repo. The link is in the description :)

  • @2122345able
    @2122345able 9 місяців тому

    where is the part where gradient decent is being coded , how the code will know when to stop ?

  • @tiagosilva856
    @tiagosilva856 2 роки тому +1

    Where did you put the "2" of the mathematical formula in dw and db on phyton?

    • @mgreek31
      @mgreek31 2 роки тому +1

      the 2 is a scaling factor that can be omitted.

    • @tiagosilva856
      @tiagosilva856 2 роки тому +1

      @@mgreek31 ty. I understand. But omitting that don't change the performance? M.S.E still be the same if we don't ommite the "2"?

    • @mgreek31
      @mgreek31 2 роки тому +1

      @@tiagosilva856 MSE will still be the same. intuitively if you multiply the 2 in the formula it scales the x for all values of x, therefore removing it will affect the whole dataset in the same way as if nothing happened

  • @MartinHolguin-d3k
    @MartinHolguin-d3k Рік тому

    your explanation is amazing and I see how linear regression works however this only works with 1 feature
    and if you want to implement it with more than one it will fail.

  • @traagshorts4617
    @traagshorts4617 2 роки тому

    How did you fit the dimensions for the test and weights in the predict function

  • @alexisadr
    @alexisadr Рік тому

    Thank you!

  • @akifarslan6022
    @akifarslan6022 Рік тому +2

    Güzel video

  • @سعیداحمدی-ل1ث
    @سعیداحمدی-ل1ث Рік тому

    لطفا ویدیو های بیشتری بسازید

  • @surendrakakani7014
    @surendrakakani7014 Рік тому

    I have taken input(200 ,1) while executing above program i am getting y_pred as (200,200) can and geting dw shape as (1,200) but dw should be (1,1) right any body explaing is that correct

  • @omaramil5103
    @omaramil5103 Рік тому

    Thank you ❤

  • @gechimmchanel2917
    @gechimmchanel2917 Рік тому +1

    Has anyone encountered the same situation as me when given a larger dataset with about 4 weights and 200 rows, the result predicts -inf or NaN , anyone have a way to fix this??

    • @gechimmchanel2917
      @gechimmchanel2917 Рік тому

      [-inf -inf -inf] -inf
      [-inf -inf -inf] -inf
      [-inf -inf -inf] -inf
      result của weight and bias :))

  • @shivamdubey4783
    @shivamdubey4783 2 роки тому

    can you plzz explain why we took x.T transpose

    • @0xmatriksh
      @0xmatriksh 2 роки тому +2

      It is because we are doing dot product of two matrices here. So we need two matrices to be in the form of mxn and nxp, that means the number of columns in first matrix should be same as the number of rows in second matrix.
      And here,
      Suppose number of rows of X is n(which is same as y)
      But the n is number of rows for both of them so we transpose X to make n in column to match the n of mxn and nxp(like explained above) to successfully dot product them

  • @prigithjoseph7018
    @prigithjoseph7018 2 роки тому

    Hi can you please upload, presentation file also

  • @Salma30b
    @Salma30b 26 днів тому

    Why don't our teachers teach us like this?? Like 15 minutes of this video is all it took to understand hours of lecturing

  • @jinxiong2973
    @jinxiong2973 2 роки тому

    how to improve my attention span , I have good ideas and soft that I tNice tutorialnk up , the problem is putting it down in fruit loops and knowing

  • @ge_song5
    @ge_song5 4 місяці тому +1

    what's her name?

  • @timuk2008
    @timuk2008 3 місяці тому

    J'(m,b) = [df / dm df/db] ===> m is the w

  • @hiteshmistry413
    @hiteshmistry413 Рік тому

    Thank you very much, You made Linear Regression very easy for me. Here is how the linear regression training looks like in action "ua-cam.com/video/2QqLl_wpfSo/v-deo.html"

  • @hannukoistinen5329
    @hannukoistinen5329 Рік тому

    Forget Python!!! It's just fashionable right now. R is much, much better!!!

  • @saideeppandey3788
    @saideeppandey3788 2 місяці тому

    Pls note x.wtranspose because w is (1,n)