Classify ship types using machine learning (beta)
This app uses a machine learning model trained with pictures of 8 different ship types:
- Passenger Vessel
- Container Vessel
- Bulk Vessel
- Roll-on/Roll-off Vessel
- Naval Vessel
- Chemical Tanker
- Sailing Ship
- Submarine
Each learning set contained between 50 and 78 images. The model itself was trained for about 1 minute.
To train the model I used Teachable Machine, an AI Experiment by Google.
Ship Classifier uses ml5, a high-level wrapper around Tensorflow.js
If you want to try using the model in your own app, it's available here.
In case you would like to train the model yourself, here you can download learning sets with sample photos.
As I mentioned earlier, to train the model I used Teachable Machine with the following configuration:
Epochs: 150
Batch Size: 16
Learning Rate: 0.00135
Hosted with now Δ
# Install dependencies
$ npm install
# Start in development mode
$ npm run dev
# Build for production
$ npm run build
MIT