4.6.2: Multi-layer perceptrons for classification - Implementing a classifier in TensorFlow.js

Поділитися
Вставка
  • Опубліковано 13 січ 2025

КОМЕНТАРІ • 12

  • @GoogleDevelopers
    @GoogleDevelopers  Рік тому +2

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

  • @adityatiwari306
    @adityatiwari306 Рік тому +5

    7:45 i get an error saying:
    caught (in promise) ReferenceError: logProgress is not defined

    • @terokykun
      @terokykun Рік тому +1

      too

    • @adityatiwari306
      @adityatiwari306 Рік тому +6

      @@terokykun
      function logProgress(epoch, logs) {
      console.log('Data for epoch ' + epoch, logs);
      }
      paste this function just after the train function. it will fix it. it seems that he forgot to show that part in the video.

  • @NikosKatsikanis
    @NikosKatsikanis Рік тому +1

    thx for your awesome support in the discord

  • @jezza735
    @jezza735 3 місяці тому

    The function `logProgress` is specified for `onEpochEnd` but it is never defined in the slides, frustrating stuff.
    As mentioned elsewhere, the source code should have been available to download so these type of issues that arise from tutorial videos can be be resolved.
    Edit: The link to the completed project is actually right at the end of the video - may have been better to include this at the beginning of the video or in the video description as the errors are encountered before the end of the video.

  • @imtanuki4106
    @imtanuki4106 Рік тому +3

    Hey - just saying - source code link should be in the notes...that's not an unreasonable request.

  • @Skylla54
    @Skylla54 5 місяців тому

    Ok he showed the link 18:05

  • @SphereofTime
    @SphereofTime 6 місяців тому

    0:41

  • @maumenvi6907
    @maumenvi6907 7 місяців тому

    I stuck in this part. I getting the error
    var src Backend = info.backend; ^
    TypeError: Cannot read properties of undefined (reading 'backend')
    Is related to
    const outputTensor = tf.oneHot(tf.tensor1d(output, 'int32'), 10)
    I am following this curse using node.js localy in my computer.
    All examples before work just fine.
    Someone has this error? Or know someting to solve this?

  • @SphereofTime
    @SphereofTime 6 місяців тому

    0:10

  • @SphereofTime
    @SphereofTime 6 місяців тому

    4:00