How to Implement Logistic Regression in Python From Scratch

Поделиться
HTML-код
  • Опубликовано: 31 окт 2021
  • The Notebook: colab.research.google.com/dri...
    Thank you for watching the video! You can learn data science FASTER at mlnow.ai!
    Master Python at mlnow.ai/course-material/python/!
    Learn SQL & Relational Databases at mlnow.ai/course-material/sql/!
    Learn NumPy, Pandas, and Python for Data Science at mlnow.ai/course-material/data...!
    Become a Machine Learning Expert at mlnow.ai/course-material/ml/!
    Don't forget to subscribe if you enjoyed the video :D

Комментарии • 26

  • @GregHogg
    @GregHogg  11 месяцев назад

    Take my courses at mlnow.ai/!

  • @ashleyb5849
    @ashleyb5849 2 года назад +5

    Was waiting for this one, thanks!!

    • @GregHogg
      @GregHogg  2 года назад +2

      Well, you no longer have to! 😃

  • @arsheyajain7055
    @arsheyajain7055 2 года назад +3

    Awesome!!!

  • @pogiribhanuteja4650
    @pogiribhanuteja4650 2 года назад +1

    I got this at the right time.

  • @commonsense1019
    @commonsense1019 2 года назад

    plzz continue this series

  • @waliddjema5442
    @waliddjema5442 2 года назад +2

    Hey mate, I am actually cursed and keep missing your lives do you mind if I ask you here?
    Or would you like me to wait for the next live?

    • @GregHogg
      @GregHogg  2 года назад

      Haha no problem; feel free to ask a question here.

    • @waliddjema5442
      @waliddjema5442 2 года назад

      @@GregHogg hey mate, thanks for this opportunity!
      First of all, I hope that you're doing great!
      secondly,
      This question might take some time, and I think that there are other people like me.
      background: non-math/stat/anything that is worth it related 'biology'
      programming skills: zero (I just started learning python)
      Goal: basically make money by working online.
      My question is, is it worth it learning data analytics/science if my only purpose is to make money as a freelance or remote worker?
      What will separate me from millions of other data scientists/analysts (by the time I finish learning, I would have a "meh" portfolio) and I am sure that there would be at least 100.000 other people in India that are better than me (I learn from them lol) so how can I compete?
      I know that nationality plays a big role in landing a remote job (seen it with my own eyes xd) and mine kind of sucks 'even traveling abroad is out of the question since I have a weak passport.'
      Thanks for reading this!

  • @hannukoistinen5329
    @hannukoistinen5329 Год назад

    For all viewers I suggest R in data-analytics! Not Python!!

  • @Mohith7548
    @Mohith7548 2 года назад

    27:00 in accuracy score, first arg is y_actual, 2nd arg is y_pred.
    You passed the other way round. LMAO

    • @GregHogg
      @GregHogg  2 года назад +1

      The order of these two doesn't matter :)

    • @tiagosilva856
      @tiagosilva856 Год назад

      @@GregHogg doesn't matter in this example cuz phyton can understand it. But it really matters in reality. The functions arguments should be called in the right order.

  • @commonsense1019
    @commonsense1019 2 года назад

    plzz use dark mode if possible

  • @tiagosilva856
    @tiagosilva856 Год назад

    The video is very good. Although he didnt have the mathematical knowledge to explain what the coeficient and the intercept means in a logistic regression. As a senior data scientist i give this free for youtube community. the coeficient tell us how much the probability of having heart disease increase or decrease when we go up in x by 1.0 value (max HR in this example ). In this case we can see that for every 1.0 increase in MaxHR the probability of having Heart Disease decrease about 3%. The intercept tell us the probability of having heart disease when the x=0 (maxHR=0), ofc in this case dont make any sense, because if u have 0 heart rate you would be dead. Although the model says you have 500% probability of having heart disease. Maths its the base to understand ML phyton and modules are tools. You can never make a proper analize of an ML output if you dont understand maths.

    • @GregHogg
      @GregHogg  Год назад

      No.

    • @tiagosilva856
      @tiagosilva856 Год назад

      @@GregHogg yes

    • @GregHogg
      @GregHogg  Год назад +1

      @@tiagosilva856 there is a 600% probability you're incorrect

    • @tiagosilva856
      @tiagosilva856 Год назад

      @@GregHogg bro i don't remember the values. You get the picture here.

    • @GregHogg
      @GregHogg  Год назад

      @@tiagosilva856 Yep, I think I have a pretty good picture of what's going on here