• Python Programming experience and libraries such as NumPy & Pandas, conceptual understanding of machine learning will be nice but not required.

Course Outline:

• 1. Machine Learning brush up
     1. What is machine learning?
     2. Machine learning techniques
     3. Evaluating machine learning models (performance metrics)
     4. Lab: Multiple labs/exercises
  2. PyTorch Fundamentals
     1. What and Why PyTorch?
     2. PyTorch Tensors
     3. Dealing with Tensor Data Types
     4. Tensor Operations
     5. Working with Tensor shapes
     6. Pegging Tensors to device
     7. Lab: Working with PyTorch
  3. Artificial Neural Networks with PyTorch
     1. Creating a PyTorch workflow to build ANN
     2. Build a PyTorch model for Linear Regression
     3. Setting up Optimizer & Loss Function
     4. Saving and Loading model
     5. Lab: ANN Regression
  4. Build ANN: Classification with PyTorch
     1. Building an ANN for Classification
     2. Learn about loss functions, metrics, and optimizers
     3. Visualizing the model predictions
     4. Understanding a multi-class classification problem
     5. Lab: ANN with classification (binary & multi-class)
  5. Computer Vision with PyTorch
     1. Understand the basics of Convolutional Neural Networks (CNN)
     2. Using a data loader & mini batch
     3. Visualizing random samples of data
     4. Setting up Device-Agnostic Code – using GPU
     5. Saving and loading best performing model
     6. Lab: Classifying images with PyTorch
  6. Writing modular code with PyTorch
     1. What is modular code?
     2. Writing Python script for PyTorch data loaders
     3. Turn model code into Python Script
     4. Lab: Modularizing ANN code
  7. Transfer Learning with PyTorch
     1. What and Why Transfer Learning?
     2. Finding pretrained models
     3. Freezing base layers for image classification problem
     4. Using feature extractor model
     5. Exploring model metrics
     6. Lab: Image classification with Transfer Learning
  8. Recurrent Neural Networks with PyTorch
     1. Understanding recurrent neural networks (RNNs)
     2. What and why RNN batches?
     3. RNN layers: LSTM, GRU
     4. Build a RNN using PyTorch on Time Series data
     5. Lab: Build RNN for time series data
  9. PyTorch: Model Deployment
     1. Understanding model deployment
     2. Build a couple of models and deploy them
     3. Explore different ways of using the deployed models
     4. Lab: Saving and Loading models in applications
  10. PyTorch: Transformers
      1. What are transformers?
      2. Why have the revolutionized AI in the past few years
      3. Understanding transformer architecture
      4. Lab: Working with Transformers

  Training material provided: Yes (Digital format)

  Hands-on Lab: Use Google’s Colab to perform the hands-on labs, also will provide instructions to run on local machine