Visão Geral

Este Curso Artificial Intelligence Deep Learning Including Generative AI Models, ensina os participantes como usar a linguagem de programação Python para construir aplicativos modernos de aprendizado de máquina (ML) que incorporam as mais recentes tecnologias de ML, como IA generativa, aprendizado profundo, processamento de linguagem natural, e visão computacional.

Os alunos são apresentados aos conceitos básicos de Python, como variáveis, tipos de dados, funções e fluxo de controle. Eles também aprendem como usar o ambiente de computação Anaconda, que vem com muitas ferramentas valiosas para ciência de dados.

Objetivo

Após realziare este Curso Artificial Intelligence Deep Learning Including Generative AI Models, você será capaz de:

• Entenda os fundamentos do aprendizado de máquina
• Preparar dados para aprendizado de máquina
• Crie e avalie modelos de aprendizado de máquina
• Aplique o aprendizado de máquina a problemas do mundo real
• Explore as últimas tendências em aprendizado de máquina
• Revise os principais conceitos do Python
• Use o ambiente de computação Anaconda
• Importe e manipule dados com Pandas
• Realize análises exploratórias de dados com Pandas e Seaborn
• Entenda as Redes Neurais Artificiais (RNAs) e o aprendizado profundo

Pre-Requisitos

Todos os participantes devem ter experiência prévia no uso de Python para realizar análises exploratórias de dados e desenvolver modelos preditivos usando técnicas de aprendizado de máquina. Os alunos também devem ter familiaridade com a codificação básica do Python.

Materiais

Conteúdo Programatico

Introduction

Review of Core Python Concepts (**if needed – depends on tool context**)

1. Anaconda Computing Environment
2. Importing and manipulating Data with Pandas
3. Exploratory Data Analysis with Pandas and Seaborn
4. NumPy ndarrays versus Pandas Dataframes

Overview of Machine Learning/Deep Learning

1. Developing predictive models with ML
2. How Deep Learning techniques have extended ML
3. Use cases and models for ML and Deep Learning

Hands-on Introduction to Artificial Neural Networks (ANNs) and Deep Learning

1. Components of Neural Network Architecture
2. Evaluate Neural Network Fit on a Known Function
3. Define and Monitor Convergence of a Neural Network
4. Hyperparameter tuning
5. Evaluating Models
6. Scoring New Datasets with a Model

Using Deep Learning for Prediction Models

Hands-on Deep Learning Model Construction for Prediction Models

1. Preprocessing Tabular Datasets for Deep Learning Workflows
2. Data Validation Strategies
3. Architecture Modifications to Managing Over-fitting
4. Regularization Strategies
5. Deep Learning Classification Model example
6. Deep Learning Regression Model example

Extending Deep Learning Models to more complex (heterogenous) data inputs

1. What happens if we do not have a rectangle of data as the input?
2. Pre-processing sequence data (i.e., time series) to use as inputs to feed-forward ANN
3. Exploring model architectures that can handle sequence data
   1. Recurrent Neural Network (RNN)
   2. Long Short Term Memory (LSTM)
   3. Transformers
4. Extending model architecture to handle heterogenous (Sequence and non-sequence) data

Natural Language Processing with Deep Learning

1. Common use cases for text data and deep learning
2. Exploratory Data Analysis on text data
3. Cleaning/pre-processing text data
4. Understanding word embeddings
5. Text Classification models
   1. Bag of Words approach
   2. RNN / LSTM modeling approaches
6. Transfer learning with text classification models: using BERT
   1. Using Hugging Face to start with state-of-the-science models
   2. Fine-tuning the model on your datasets

Computer Vision with Deep Learning

1. Common AI use cases with images
2. Exploratory Data Analysis on image data
3. Pre-processing images
4. Data augmentation with existing images
5. Image classification examples
   1. Image classification with ANN
   2. Image classification with convolutional neural networks
6. Image classification and transfer learning:
   1. Using Hugging Face to start with state-of-the-science models
   2. Fine-tuning the model on your datasets
7. Image segmentation and transfer learning
   1. Using Hugging Face to start with state-of-the-science models
   2. Fine-tuning the model on your datasets

Generative AI with Deep Learning

• Generative AI fundamentals
  1. Generating new content versus analyzing existing content
  2. Example use cases: text, music, artwork, code generation
  3. Ethics of generative AI
• Sequence Generation with RNN
  1. Recurrent neural networks overview
  2. Preparing text data
  3. Setting up training samples and outputs
  4. Model training with batching
  5. Generating text from a trained model
  6. Pros and cons of sequential generation
• Overview of current popular large language models (LLM)
  1. ChatGPT
  2. DALL-E 2
  3. Bing AI
• Medium-sized LLM in your environment
  1. Stanford Alpaca
  2. Facebook Llama
  3. Transfer learning with your data in these contexts