3.2: Selecting an ML model to use

แชร์
ฝัง
  • เผยแพร่เมื่อ 16 ก.ย. 2024
  • Now you know what pre-trained models are, how do you select one to use in a real world project? In this section you will walk through a hypothetical real world scenario, breaking down the problem into the following steps:
    1. What exactly is the problem you are trying to solve?
    2. Once you know the problem, what models do you know of that could assist you in solving this problem?
    3. Considering all identified usable models, how do you select one to deliver the best solution?
    Here we will cover concepts such as model size (both file size and RAM execution size), and model performance, along with other criteria to find the best model to choose.
    Catch more episodes from Machine Learning for Web Developers (Web ML) → goo.gle/learn-...
    Check out TensorFlow on TH-cam → goo.gle/Tensor...
    Subscribe to Google Developers → goo.gle/develo...
    Connect with Jason Mayes to ask questions:
    LinkedIn → goo.gle/3GwgeLw
    Twitter →goo.gle/3Xh6MT7
    Discord →goo.gle/3WWVO5t
    Use #WebML to share your learnings and creations from this course to meet your peers on social media!
    See what others have already made with Web ML → goo.gle/made-wi...

ความคิดเห็น • 13

  • @GoogleDevelopers
    @GoogleDevelopers  ปีที่แล้ว +2

    Catch more episodes from Machine Learning for Web Developers (Web ML) → goo.gle/learn-WebML

  • @Gabriel-iq6ug
    @Gabriel-iq6ug ปีที่แล้ว +5

    All the videos in this playlist surprised me by teaching me something I didn't expect. This feels great

    • @JasonMayes
      @JasonMayes 8 หลายเดือนก่อน +2

      That's great to hear!

  • @ErickWendelTraining
    @ErickWendelTraining ปีที่แล้ว +3

    I've been enjoying a lot this course! Thanks Jason for the amazing work

    • @JasonMayes
      @JasonMayes ปีที่แล้ว +1

      Most welcome Erick! Thanks for the kind words!

  • @himnishbhansali2099
    @himnishbhansali2099 7 หลายเดือนก่อน

    Amazing course I've been using AI and this course has helped me a lot, thanks jason and google team.

  • @aneerimmco
    @aneerimmco 2 หลายเดือนก่อน

    informative!!!

  • @Gabriel-iq6ug
    @Gabriel-iq6ug ปีที่แล้ว +2

    With WebGPU (enabled in Chrome by default recently), I would expect the FPS to get higher, but I haven't tried it yet

    • @jasonmayes_goog
      @jasonmayes_goog 11 หลายเดือนก่อน

      It can run models around 2 - 3.8x faster depending on the model architecture

  • @RikyPerdana
    @RikyPerdana ปีที่แล้ว

    Runtime memory is none of my concern. I can download more RAM whenever I need to

  • @BorhanChowdhuryChittagong
    @BorhanChowdhuryChittagong ปีที่แล้ว +2

    your ways of breaking down a problem and picking the right solution for the problem is amazing. Really appreciate your effort to teaching us ML :) @jason mayes

    • @jasonmayes_goog
      @jasonmayes_goog 11 หลายเดือนก่อน

      Glad you are enjoying! Sorry for not replying sooner as this is hosted on Google Developers I do not get notifications on comments.