MultiClass Logistic Regression, Python Machine Learning, Detecting multi outcome logistic regression

แชร์
ฝัง
  • เผยแพร่เมื่อ 21 ธ.ค. 2024
  • In this #PythonMachineLearning series, #MultiClassLogisticRegression is explained step by step using #IRISDataset. Logistic regression is applied on iris dataset and the species of iris flower is detected using its #PetalLength, #PetalWidth, #SepalLength, #SepalWidth.
    Link to the code:
    github.com/kjs...
    Link to Iris dataset used:
    www.kaggle.com...
    Link to Machine Learning playlist:
    • Machine Learning with ...
    Link to Chatbot playlist:
    • Build your own Chatbot...
    Link to Python playlist:
    • Python Basics Tutorial...
    Link to Pandas playlist:
    • Python Pandas Tutorial...
    Link to NumPy playlist:
    • NumPy Playlist || NumP...
    You can connect with me here:
    / atomai.online
    / atomai_online

ความคิดเห็น • 14

  • @sohailshaikh8571
    @sohailshaikh8571 2 ปีที่แล้ว +4

    Very good approach... One thing could have be applied here for converting Species into numeric data is, LabelEncoder. It will be flexible for any number of species.

  • @sebastiendupre7098
    @sebastiendupre7098 ปีที่แล้ว +1

    lovely content

  • @jw1984mouse
    @jw1984mouse 3 ปีที่แล้ว +1

    Thank you for this video. It helps me a lot with my project.

  • @lucillewiid5476
    @lucillewiid5476 2 ปีที่แล้ว

    The best tutorial thanks!

  • @raj345to
    @raj345to 3 ปีที่แล้ว

    very nicely explained.

  • @MenSpace-w6p
    @MenSpace-w6p 2 ปีที่แล้ว

    How to extract sigmoid function from it

  • @yangfarhana3660
    @yangfarhana3660 3 ปีที่แล้ว +1

    Thank you for this tutorial, it's very clear!
    I have a question on the final part of the video, for the same model, with a test size of 0.3, how we improve the score of 0.933, what do we have to tweak?

    • @raj345to
      @raj345to 3 ปีที่แล้ว

      i think normalizing/scaling the data and hyper parameter tuning may help

  • @ifyifemanima3972
    @ifyifemanima3972 ปีที่แล้ว

    Thank you!

  • @gokulgupta1021
    @gokulgupta1021 3 ปีที่แล้ว +1

    Can you make video without using sklearn. Using sklearn a baby would make a model. A video should have a proper knowledge that gradient with batch etc.

  • @tiagosilva856
    @tiagosilva856 2 ปีที่แล้ว

    And BTW you should not label y as 1,2,3 but 0,1,2 to proper analyze the confusion matriz

  • @tiagosilva856
    @tiagosilva856 2 ปีที่แล้ว

    Use label encoder.

  • @tiagosilva856
    @tiagosilva856 2 ปีที่แล้ว

    Wrong. U don't use score in logistic regressions. U use accuracy.

  • @tiagosilva856
    @tiagosilva856 2 ปีที่แล้ว +1

    Don´t use replace its a bad praticse. Use catergory instead:
    df['Species']=df['Species'].astype('category')
    df['Species']=df['Species'].cat.codes
    df