Replication Data for: A robotic assistant for pictogram classification in education: a proposal using dynamically quantized deep neural networks and the CIFAR-10 dataset for developing countries

This dataset contains the source code (Jupyter Notebooks) and experimental scripts developed for the research article "A robotic assistant for pictogram classification in education". The study presents a low-cost robotic assistant designed to support pictogram recognition for children (ages 5-8) in developing countries, specifically Ecuador. The repository includes: 1. Deployment Code: Scripts to control the robotic assistant, handle webcam image capture, and perform real-time classification. 2. Training/Validation Code: Scripts used to fine-tune and evaluate the mobilenetv3_small neural network, specifically focusing on dynamic quantization to optimize performance on low-resource hardware. (2026-01-10)