Teachable Lab Task

Activity: Creating an Image Classification Model with Teachable Machine

Objective: To prepare a small dataset and train an image classification model to categorize five distinct objects using Teachable Machine.

Tool Required:

• Teachable Machine: https://teachablemachine.withgoogle.com/ (Web-based)

---

Step 1: Prepare the Image Dataset

First, you need to collect images for each category. For this exercise, you must have at least 10-20 images for each of the following five categories:

1. books (Images of physical books, stacks of books, or libraries.)

2. computers (Images of desktop computers, monitors, laptops, or tablets.)

3. happy human face (Images of people clearly showing a happy or smiling face.)

4. watches (Images of wristwatches (analog or digital) and smartwatches.)

5. nature (Images of landscapes, forests, mountains, rivers, or plants.)

---

Step 2: Create the Image Project in Teachable Machine

1. Open Teachable Machine in your browser: https://teachablemachine.withgoogle.com/

2. Click "Get Started".

3. Select "Image Project".

4. Choose the "Standard image model" option.

Step 3: Define the Classes and Upload Data

You will see two default classes. You need to rename these and add three more to match your five categories.

1. Rename Class 1: Click the pencil icon next to "Class 1" and rename it to books.

2. Upload Data for Class 1:

   • Click the "Upload" button under the books class.

   • Select "Choose images from your files" and upload all your images collected for the books category.

3. Repeat for other classes:

   • Rename "Class 2" to computers and upload the corresponding images.

   • Click "Add a class" and rename the new class to happy human face and upload images.

   • Click "Add a class" and rename the new class to watches and upload images.

   • Click "Add a class" and rename the new class to nature and upload images.

Step 4: Test the Model

Test the model by providing the below images as input to it and write your responses 

image1	image2	image3

 

Observation:

Image	Happy (%)	Sad (%)	Surprise (%)