ONNX
Open Neural Network Exchange — an open format for representing machine learning models that enables interoperability between different ML frameworks and deployment targets.
Why It Matters
ONNX solves the portability problem — train in PyTorch, deploy in TensorFlow or on specialized hardware, without rewriting model code.
Example
Training a model in PyTorch, exporting it to ONNX format, then deploying it on an edge device using ONNX Runtime for optimized inference.
Think of it like...
Like PDF for documents — a universal format that preserves the content regardless of which tool created it or which tool opens it.
Related Terms
Model Serving
The infrastructure and process of deploying trained ML models to production where they can receive requests and return predictions in real time. It includes scaling, load balancing, and version management.
Deployment
The process of making a trained ML model available for use in production applications. Deployment involves packaging the model, setting up serving infrastructure, and establishing monitoring.