TensorFlow 2.0 – Pedro Lelis

TensorFlow 2.0 - Pedro Lelis

Existem várias alterações no TensorFlow 2.0 para tornar os usuários do TensorFlow mais produtivos. O TensorFlow 2.0 remove APIs redundantes, torna as APIs mais consistentes (RNNs unificadas, otimizadores unificados) e se integra melhor ao tempo de execução do Python com a Eager execution. Este guia apresenta uma visão de como deve ser o desenvolvimento no TensorFlow 2.0.

Moved by curiosity, I’m an entrepreneurial scientist who seeks to improve people’s lives through Artificial Intelligence solutions. I’m data scientist at CI&T, dean at School of AI and Google Cloud Certified both as Professional Data Engineer and as Professional Cloud Architect. I’m also speaker, coordinator and organizer in academic and business conferences. Specialties: Python, TensorFlow, Google Cloud and Teaching (People and Machines). Main researches and interests: Fairness in Machine Learning, Few-shot Learning, Memory-Augmented Neural Network, Transformer (Attention), Generative Adversarial Network and Reinforcement Learning.

Deixe uma resposta

O seu endereço de e-mail não será publicado. Campos obrigatórios são marcados com *