STM32N6 NPU Deployment — Politecnico di Milano  1.0
Documentation for Neural Network Deployment on STM32N6 NPU - Politecnico di Milano 2024-2025
Python Deployment Pipeline

Python scripts orchestrating the full model-to-firmware pipeline.

Python scripts orchestrating the full model-to-firmware pipeline.

This module group collects all Python source files that implement the ModelZoo Services deployment pipeline, from YAML configuration parsing to C code generation and firmware flashing.

Pipeline execution order:

user_config.yaml
stm32ai_main.py ← Entry point, dispatches operation mode
├── quantize.py ← INT8 PTQ via TFLite Converter
├── preprocess.py ← Dataset loading and normalization
├── postprocess.py ← Heatmap / NMS decoder
├── evaluate.py ← OKS/AP metrics
└── deploy.py ← Deployment orchestrator
├── gen_h_file.py ← Generates app_config.h (Python→C bridge)
└── common_deploy.py ← ST Edge AI Core + STM32CubeIDE invocation
└── external_memory_mgt.py ← C source patching
Definition: deploy.py:1