NeuronetworksAI Ed

Home / Projects / Image classification

Year 2 · AI/ML Engineering

Imageclassificationapplication

Build a computer vision application that classifies images using deep learning. Train a model on a dataset, evaluate its performance and create a simple web interface to try it yourself.

  • Python
  • PyTorch
  • Computer Vision
  • Deep Learning
Image ClassificationIllustrative demo output

Upload an image

Photo of a golden retriever used as the input image in the image classification demoChoose an imageor drag and drop JPG, PNG

Prediction results

  • Golden Retriever92.4%
  • Labrador Retriever4.2%
  • German Shepherd1.8%
  • Other1.6%

Model information

Model
ResNet18 (pre-trained)
Input size
224 × 224
Classes
1000 (ImageNet)
Golden Retriever · 92.4%Top prediction
ResNet18224 × 224 input

Project overview

What you build

A complete image classification application that takes an image as input, runs it through a trained deep learning model and shows the top predictions with confidence scores in a clean web interface.

NeuronetwrksIllustrative demo output

Image Classification

  • Predict
  • Model info
  • History

Upload an image

Drag and drop an image here
or click to browse
Photo of a golden retriever used as the input image in the image classification demo

Top predictions

  • 1Golden Retriever92.4%
  • 2Labrador Retriever4.2%
  • 3German Shepherd1.8%
  • 4Other1.6%
Run prediction
  • Upload an imageDrag and drop an image or choose a file from your device.
  • Run a trained modelLoad a pre-trained model and run inference on your image.
  • View predictionsSee the top predicted classes with confidence scores.
  • Explore model detailsCheck model information, input size and the set of classes.

From data to deployment

How it works

Follow a clear, end-to-end process to build the image classification application, from preparing data to deploying a simple web interface.

  1. 01Prepare datasetUse a labelled dataset such as ImageNet, or a smaller dataset provided in the program.
  2. 02Train or load modelTrain a model or use a pre-trained model with transfer learning.
  3. 03Evaluate predictionsTest the model on unseen images and review the results.
  4. 04Deploy interfaceBuild a simple web app with Streamlit to try it yourself.

A closer look

Inside the application

A clean and intuitive interface to upload images, view predictions and explore how the model performs.

  • Upload & predict
  • Top predictions
  • Model information
  • Example images
Image ClassificationIllustrative demo output

Input image

Photo of a golden retriever used as the input image in the image classification demo
Change imagegolden_retriever.jpg
224 × 224

Prediction results

  • 1Golden Retriever92.4%
  • 2Labrador Retriever4.2%
  • 3German Shepherd1.8%
  • 4Other1.6%
ModelResNet18 (pre-trained)
Input size224 × 224
Number of classes1000 (ImageNet)

Tech stack

Tools & technologies

Use industry-standard tools and libraries for computer vision and deep learning.

  • PythonCore programming language
  • PyTorchDeep learning framework
  • torchvisionVision models and datasets
  • StreamlitBuild the web interface

Key learning outcomes

Skills you practise

Build practical skills that you can apply to real-world computer vision problems.

  • Data preparationWork with image datasets and understand data preprocessing steps.
  • Model training and inferenceTrain a model or use a pre-trained model with transfer learning.
  • Model evaluationMeasure performance and interpret prediction results.
  • Building a web interfaceCreate a simple application using Streamlit.

Project roadmap

Part of the AI/ML Engineering program

This project is part of the Year 2 curriculum in the AI/ML Engineering three-year program. It builds on core machine learning concepts and helps you apply them in a practical application.

AI/ML EngineeringYear 2 · Core project

Your next step

Ready to explore more projects?

Discover more hands-on projects from the AI/ML Engineering program and build a portfolio you can show.