Python Feature Scaling in SciKit-Learn (Normalization vs Standardization)

Поділитися
Вставка
  • Опубліковано 5 січ 2025

КОМЕНТАРІ • 30

  • @RyanAndMattDataScience
    @RyanAndMattDataScience  5 місяців тому

    Hey guys I hope you enjoyed the video! If you did please subscribe to the channel!
    Join our Data Science Discord Here: discord.com/invite/F7dxbvHUhg
    If you want to watch a full course on Machine Learning check out Datacamp: datacamp.pxf.io/XYD7Qg
    Want to solve Python data interview questions: stratascratch.com/?via=ryan
    I'm also open to freelance data projects. Hit me up at ryannolandata@gmail.com
    *Both Datacamp and Stratascratch are affiliate links.

  • @AktamNarzullayev-f7m
    @AktamNarzullayev-f7m 9 місяців тому +7

    underrated channel great video

  • @v.jananayagan3284
    @v.jananayagan3284 7 місяців тому +1

    you teach very well than other channels but i don't know why pepoles are not spend time on your channel really helpfull man

  • @danieljuniormilazi7701
    @danieljuniormilazi7701 5 місяців тому

    Dude you just made the whole concept so easy to understand, i've been trying to understand exactly what was required of me for hours. Keep up the great work ❣❣❣❣❣❣

  • @morsiahinimdou9218
    @morsiahinimdou9218 2 місяці тому

    thank you good bless you i think after all these video i'll understand so well the machine learning

  • @photonganglol2413
    @photonganglol2413 3 місяці тому +2

    as someone who is new to AI/ML, maybe some more clear terminology defined would be helpful. A lot of resources call what you describe as 'Normalizing' as 'Scaling'. And what you call 'standardization' is referred to as 'Normalizing'. Just a little confusing but great video actually showing the difference between the 2.

  • @sandeep-kc9hs
    @sandeep-kc9hs 6 місяців тому

    learned a lot from this. excellent teaching🙌

  • @lancerkind
    @lancerkind 10 місяців тому

    Very good video! I learned a lot. If I was to ask for more, it would be to fill in WHY normalize or standardized. You mention some about “getting your numbers in order.” Add to that there are reasons for visualization tools, comparison analysis, and whatever else. I have some ideas why, but I’m guessing as a Pandas user you have encountered many more.
    Thank you for sharing.

    • @RyanAndMattDataScience
      @RyanAndMattDataScience  10 місяців тому

      No problem and I may make a statistics course video in the future. Just waiting on my job to apply more skills

  • @Welcomereddy
    @Welcomereddy Рік тому +1

    Excellent brother !

  • @ShiftKoncepts
    @ShiftKoncepts 21 день тому

    is there any back transformation taht needs to be done afterwards?

  • @ShiftKoncepts
    @ShiftKoncepts 22 дні тому

    Should I do polynomial and/or log transformations before normalizing or after?

  • @onurdatascience
    @onurdatascience Рік тому +1

    Great video!

  • @lilikoimahalo
    @lilikoimahalo Рік тому

    Could you also explain how the choice of feature_range affects the output processing please? Trying to understand in which case it should be (0,5) and when it should be (0,10), and how you then interpret the output, for example? Also, I am wondering: you are applying scalers to the whole dataset, but what if you have a regression type task (predicting an actual number)? If you apply scalers to all columns then your targets also change

  • @darkredrose7683
    @darkredrose7683 Місяць тому

    Would it make sense to do a kruskal-wallis significance test for scaled indices that have been scaled 0-1 with min-max? Thank you ❤

    • @RyanAndMattDataScience
      @RyanAndMattDataScience  Місяць тому +1

      Actually just released that video a few weeks ago. Finishing up a stats playlist

  • @sara-sx7gm
    @sara-sx7gm 7 місяців тому

    Helpful . Thank you so much

  • @redeemmbonge
    @redeemmbonge 2 місяці тому

    👏👏👏

  • @rishikeshjadhav4774
    @rishikeshjadhav4774 6 місяців тому +1

    can you please post the jupyter notebook containing code , it will be very healpful

    • @RyanAndMattDataScience
      @RyanAndMattDataScience  6 місяців тому

      Will be on my website soon, I’m moving the code from the vids into articles

  • @qaisshefa4846
    @qaisshefa4846 6 місяців тому

    Thanks so much

  • @gnuwaves743
    @gnuwaves743 2 місяці тому

    Is there an easy way to get the column names? I have almost 100.

    • @soomann2716
      @soomann2716 27 днів тому

      df.loc[ : , [ ' Col1 ' , ' Col2 ' , ' ColN ' ]]
      if from index use df.iloc[]