Background Inquiry

First of all, we would like to get a sense of your prior experience with Machine Learning model.

<aside> ❓ How much experience do you have in training Machine Learning (ML) models?

</aside>

<aside> ➡️

</aside>

Intro to Teachable Machine

Next, we would like to get a sense of your current knowledge and intuition of a Machine Learning model. The graphics below summarizes 2 main stages of a Machine Learning modeling process:

  1. Train a model with a set of provided data

    Group 13.png

  2. Test the trained model with another set of unseen data

    Group 12.png

Pre-session Assessment

Now that you know a bit about the model training process, we will show you several model training and testing scenarios, and then ask you to predict the output from each of the model.

  1. Scenario One: A model is trained with to distinguish 2 classes: "Bill", "Wall".

    Group 14Scenario #1.png

    How would the model classify the image of a new set of background?

    Group 15.png

    <aside> 💡 Note #1: The percentage bar for the following options represents the model's confidence in classifying the input into the respective class.

    </aside>

    <aside> 💡 Note #2: If you think the model cannot make the classification, you can put 50% - 50% on the percentage bars.

    </aside>

    <aside> ✏️ Draw your answer below

    </aside>

    Group 27V2.png

    How would the model classify an image of Bill in the new set of background?

    Group 12.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 27V2.png

  1. Scenario #2: A model is trained with to distinguish 2 classes: "Bill", "ShaoBo".

    Group 16new.png

    How would the model classify an image of Bill with his face partially covered by his palm?

    Group 18V2.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 26V2.png

    How would the model classify an image of Bill with his face fully covered by his palm?

    Group 17V2.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 26V2.png

    How would the model classify an image of ShaoBo with his face covered by Bill's palm?

    Group 19new.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 26V2.png

  1. Scenario #3: A model is trained with to distinguish 3 classes: "Bill", "ShaoBo", "Bill & ShaoBo". In particular, the training dataset for class "Bill" contains image of himself with a thumb-up.

    Group 20new.png

    How would the model classify an image of ShaoBo with a thumb-up?

    Group 23new.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 28V2.png

    How would the model classify an image of Bill & ShaoBo if one of them shows a thumb-up?

    Group 21new.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 28V2.png

    How would the model classify an image of Bill & ShaoBo if both of them show a thumb-up?

    Group 22new.png

    <aside> ✏️ Draw your answer below

    </aside>

    Group 28V2.png