WizModel

What is WizModel?

WizModel is a platform that simplifies the deployment of machine learning models to production by providing a unified API for ML inference. Users can call the WizModel API to perform ML inference on any model without having to write all the necessary code for deployment and inference. It offers a production-ready solution for model deployment and provides a demo UI for users to try the model without the need for additional code.

Who created WizModel?

WizModel was created by a team that developed Cog2, a tool for packaging machine learning models in a production-ready container. The company offers a unified API for machine learning inference, simplifying deployment to production. Users can call the WizModel API for ML inference without the need to write extensive code for model deployment.

What is WizModel used for?

• Unified API for ML inference on any model
• Production-ready model deployment
• Easy ML model inference without writing excessive code or dealing with dependencies
• Demo UI for trying out models
• Cloud deployment support
• Automated model packaging in a production-ready container
• Efficient model setup and prediction processing
• Local and cloud prediction capabilities
• REST API access for inference
• Simplified AI tool for configuration generation

Who is WizModel for?

• Data scientists
• Machine learning engineers
• AI researchers
• AI/ML developers

How to use WizModel?

To use WizModel, follow these steps:

1. Define the environment by creating a new project, initializing it, and adding necessary configurations in the cog.yaml file.
2. Write the predict.py script to define how predictions are run on your model.
3. Run predictions locally using cog2 predict command and then push the model to the server with cog2 push.
4. To run the model in the cloud, call the REST API with Python using the provided URL and API key.

Remember to set up the necessary configurations, Python packages, and model prediction logic as described in the user guide.