Overview
A FastAPI inference service serving a fine-tuned GPT-2 causal model with CORS enabled, logging, and config in config.yaml. Responses come from generate_response.py and the server entry is api_server.py, with Docker and Kubernetes manifests included. The stack is FastAPI, Transformers, and PyTorch, and thread consumes it.
Installation
git clone https://github.com/coccinella-labs/mlapi