Dog Breed ClasssifAIer, powered by DataRobot Visual AI

Find a dog. Take a photo of a dog. Describe the dog. Upload to the AI 🤖🧠☁️. Learn what breed the dog is. Easy enough with DataRobot.

This is a fun little demo app showing DataRobot's Visual AI in practice. See the app in action: link to deployed app.

Learn how it's made by following a tutorial on DataRobot Developer portal.

Project was bootstrapped using Nitro.

Usage

Prerequisites

• DataRobot account. You can request a free trial at https://datarobot.com/lp/trial.
• Vercel account with Vercel CLI set up (for deployment and API) - https://vercel.com

Deploy the models

The training datasets are included and located in datasets directory. There are 5 of them, labeled x-small to x-large - based on dog weight. To use, you need to train and deploy all 5 models with DataRobot and use their deployment IDs in the environment variables (see steps below).

Install and Run Dev Server

• Set up environment variables

cp env.sample .env

• Fill in your details into the environment variables.

• Set up Vercel project for local development

vercel init

• Install the dependencies:

npm install

• Run the development server

npm run dev

Production

Locally:

  npm run build
  npm run start

On Vercel:

vercel --prod

Development and Contributing

If you'd like to report an issue or bug, suggest improvements, or contribute code to this project, please refer to CONTRIBUTING.md.

Code of Conduct

This project has adopted the Contributor Covenant for its Code of Conduct. See CODE_OF_CONDUCT.md to read it in full.

License

Licensed under the Apache License 2.0. See LICENSE to read it in full.