59 - What is Random Forest classifier?

Поділитися
Вставка
  • Опубліковано 14 жов 2019
  • There are many machine learning classifiers that have been developed over years for various purposes. While deep learning is slowly replacing these traditional classifiers, Random Forest still beats deep learning for applications with limited training data. Microscope image segmentation is one such application where users often work with limited training data. This tutorial provides a quick introduction to Random Forest - no coding!
    The code from this video is available at: github.com/bnsreenu/python_fo...
  • Наука та технологія

КОМЕНТАРІ • 18

  • @surajshah4317
    @surajshah4317 3 роки тому

    dear sir .. number of decison tree are equal to number of input features ?

  • @yontenjamtsho4315
    @yontenjamtsho4315 2 роки тому

    Is it necessary to create a bootstrap dataset?

  • @mesfinabate5952
    @mesfinabate5952 Рік тому

    How do we know the pixel value?

  • @williamcabisca5058
    @williamcabisca5058 2 роки тому

    Awesome info

  • @lee-qk2vk
    @lee-qk2vk 3 місяці тому

    so i have watched your videos from 1 upto thus far, you have not shown how to LABEL the image

  • @johnpuskin463
    @johnpuskin463 3 роки тому +2

    Thank you for your good explanations and experience based valuable suggestions. You claimed that Random Forest based segmentation is better than that of SVM. Do you have a chance to test XGBoost? If no then I strongly suggest it.

    • @DigitalSreeni
      @DigitalSreeni  3 роки тому +4

      Yes, I did experiment with a few boosting techniques including XGBoost. It is a good approach but the accuracy depends on hyperparameters. I thought I recorded a video on this topic but apparently I haven't. Thanks for the suggestion, I am sure the viewers of this channel would like to know more about it.

  • @mesfinabate5952
    @mesfinabate5952 Рік тому +1

    Sir. How to measure its performance?

    • @idrisseahamadiabdallah7669
      @idrisseahamadiabdallah7669 10 місяців тому

      I think that because RandomForest is a classification algorithm so classification metrics such as accuracy, roc&auc , confusion metrics can be used to assess the model.

  • @boooringlearning
    @boooringlearning 3 роки тому

    great video!

  • @AgriculturaDigital
    @AgriculturaDigital 4 роки тому

    Your videos is very good one. Thanks very much man. Do you give the permission to use your exemples and code to recorde a Portuguese videos?

    • @DigitalSreeni
      @DigitalSreeni  4 роки тому

      If you want to add Portuguese voice over please let me know. I guess, it can help some users.

    • @AgriculturaDigital
      @AgriculturaDigital 3 роки тому

      Python for Microscopists by Sreeni . No no...The idea is record my own video. Just use your code exemples.

  • @user-qx6ki8dh8o
    @user-qx6ki8dh8o Рік тому

    Dear. I do appreciate your lecturing style. I have two questions- "How to measure the performance of Random Forest?". What metrics are used to compare it to other algorithms?

    • @idrisseahamadiabdallah7669
      @idrisseahamadiabdallah7669 10 місяців тому

      I think that because RandomForest is a classification algorithm so classification metrics such as accuracy, roc&auc , confusion metrics can be used to assess the model.

  • @DeepIITH
    @DeepIITH Рік тому

    Can this be applied to HeLa Cell segmentation?

    • @DigitalSreeni
      @DigitalSreeni  Рік тому

      Yes. The exact application doesn't matter.