STM32N6 NPU Deployment — Politecnico di Milano  1.0
Documentation for Neural Network Deployment on STM32N6 NPU - Politecnico di Milano 2024-2025
preprocess.py File Reference

Dataset preprocessing pipeline for pose estimation. More...

Go to the source code of this file.

Namespaces

 preprocess
 

Functions

Tuple preprocess.preprocess (DictConfig cfg=None)
 
def preprocess.apply_rescaling (tf.data.Dataset dataset=None, float scale=None, float offset=None)
 
tf.Tensor preprocess.preprocess_input (np.ndarray image, dict input_details)
 

Detailed Description

Dataset preprocessing pipeline for pose estimation.

Loads, resizes, normalizes and batches images and keypoint annotations from COCO-format datasets. Returns TensorFlow Dataset objects for training, validation, quantization calibration, and test splits.

Note
Politecnico di Milano, A.Y. 2024-2025. Multidisciplinary Project — Neural Network Deployment on STM32N6 NPU.

Definition in file preprocess.py.