Visão Geral

Curso Deep Learning with Microsoft Cognitive Toolkit. O Microsoft Cognitive Toolkit (também conhecido como CNTK) permite que você aproveite a inteligência em grandes conjuntos de dados por meio de aprendizado profundo, fornecendo dimensionamento, velocidade e precisão descomprometidos com qualidade de nível comercial e compatibilidade com as linguagens de programação e algoritmos que você já usa. No Curso Deep Learning with Microsoft Cognitive Toolkit, você aprenderá sobre as aplicações de Deep Learning usando Microsoft CNTK.

Objetivo

Ao participar do Curso Deep Learning with Microsoft Cognitive Toolkit, os participantes aprenderão:

• API Python do Microsoft Cognitive Toolkit
• Classificação de imagens com rede neural profunda
• Classificação de imagens com rede neural convolucional (CNN)
• Modelagem de Sequência com Rede Neural Recorrente (RNN)
• Redes Adversariais Gerativas (GAN)

Publico Alvo

Curso Deep Learning with Microsoft Cognitive Toolkit é para analistas de dados, cientistas de dados e desenvolvedores de software

Materiais

Conteúdo Programatico

Getting Started

1. What is Microsoft Cognitive Toolkit
2. Install and Run Microsoft Cognitive Toolkit Python API

Datasets

1. MNIST Handwritten Digits Dataset
2. CIFAR Image Dataset
3. Iris Flower Dataset

Machine Learning with Microsoft Cognitive Toolkit

1. Loss Function and Optimizer
2. Linear Regression Model
3. Logistics Regression Model
4. Save and Load Model

Neural Network (NN) with Microsoft Cognitive Toolkit

1. What is Neural Network
2. Activation Functions
3. MNIST Handwritten Digit Recogition with Neural Network

Convolutional Neural Network (CNN)

1. What is Convolutional Neural Network?
2. CNN Architecture
3. MNIST Handwritten Digit Recogition with CNN
4. Transfer Learning
5. ResNET

Recurrent Neural Network (RNN)

1. What is Recurrent Neural Network?
2. Language Understanding with RNN
3. LSTM Based Time Series
4. Sequence to Sequence Modeling

Object Recognition with R-CNN

1. What is Object Recogintion?
2. What is R-CNN?
3. Fast R-CNN
4. Faster R-CNN

Generative Adversarial Network (GAN)

1. What is Generative Adversarial Network ?
2. Deep convolutional GAN (DCGAN) network