(Code) What is Winsorization | Using percentiles for capping outliers in Python | Machine Learning

Поділитися
Вставка
  • Опубліковано 20 гру 2024

КОМЕНТАРІ • 12

  • @akashmishra7616
    @akashmishra7616 4 роки тому

    hello Rachit, I was looking for this tutorial everyone but couldn't find it and Bhagwan ki Kripa se I saw your video. it was really nice. Thanks for this video buddy.

  • @gouthamansaravanan7692
    @gouthamansaravanan7692 2 роки тому

    Hi, very nice Tutorial. Thank you very much!!

  • @sodiqrafiu9072
    @sodiqrafiu9072 4 роки тому +1

    Hello Rachit, Thanks for all, we appreciate. We are still looking forward to your end to end project. Thanks once more

    • @rachittoshniwal
      @rachittoshniwal  4 роки тому +1

      Hi Sodiq, thanks for the kind words! I have that end to end project in the queue. But it might take some time tbh. Until then I highly encourage you to follow this:
      ua-cam.com/play/PLeo1K3hjS3ut2o1ay5Dqh-r1kq6ZU8W0M.html
      You'll learn a lot here! :)

    • @ajaykushwaha-je6mw
      @ajaykushwaha-je6mw 3 роки тому +1

      Yes Rachit we are looking for end to end which must involve all feature engineering technique.

  • @titusw5145
    @titusw5145 2 роки тому

    Hello Rachit, thx for your video! Do you have any clue on how one would winsorize only within a given year? Say we have a dataframe with different variables, which are sorted by year. How would we initialise the winsorisation process not for the whole dataframe but only within those groups of years?

    • @rachittoshniwal
      @rachittoshniwal  2 роки тому

      I guess you could try: df.groupby('Year').apply(some_function) and of course, this some_function will have stuff that will do the winsorization, and it will do it one year at a time.

  • @lautarocisterna3339
    @lautarocisterna3339 3 роки тому

    Hi man! Is there a way to add the percentile capping to a pipeline? Since is not a transformer we won't be able to add it like that, right?

  • @farihatahosinboishakhi9290
    @farihatahosinboishakhi9290 3 роки тому

    Thanks that was helpful

  • @srinathreddy8435
    @srinathreddy8435 3 роки тому

    Hello, Nice tutorial I appreciate your effort, do you have a full video discussing about plotting the graphs for a data sets with multiple columns simultaneously like you've did in this video. Thanks in advance.

  • @christianrodier3381
    @christianrodier3381 2 роки тому

    That was helpful