PyTorch Ultimate 2024: From Basics to Cutting-Edge

Contents

What You’ll Learn Requirements Description Who this course is for:

What
You’ll Learn

learn all relevant aspects of PyTorch from simple models to state-of-the-art models
deploy your model on-premise and to Cloud
Transformers
Natural Language Processing (NLP)
e.g. Word Embeddings
Zero-Shot Classification
Similarity Scores
CNNs (Image-
Audio-Classification; Object Detection)
Style Transfer
Recurrent Neural Networks
Autoencoders
Generative Adversarial Networks
Recommender Systems
adapt top-notch algorithms like Transformers to custom datasets
develop CNN models for image classification
object detection
Style Transfer
develop RNN models
Autoencoders
Generative Adversarial Networks
learn about new frameworks (e.g. PyTorch Lightning) and new models like OpenAI ChatGPT
use Transfer Learning

Requirements

basic Python knowledge

Description

PyTorch is a Python framework developed by Facebook to develop and deploy Deep Learning models. It is one of the most popular Deep Learning frameworks nowadays.

In this course you will learn everything that is needed for developing and applying Deep Learning models to your own data. All relevant fields like Regression, Classification, CNNs, RNNs, GANs, NLP, Recommender Systems, and many more are covered. Furthermore, state of the art models and architectures like Transformers, YOLOv7, or ChatGPT are presented.

It is important to me that you learn the underlying concepts as well as how to implement the techniques. You will be challenged to tackle problems on your own, before I present you my solution.

In my course I will teach you:

Introduction to Deep Learning
- high level understanding
- perceptrons
- layers
- activation functions
- loss functions
- optimizers
Tensor handling
- creation and specific features of tensors
- automatic gradient calculation (autograd)
Modeling introduction, incl.
- Linear Regression from scratch
- understanding PyTorch model training
- Batches
- Datasets and Dataloaders
- Hyperparameter Tuning
- saving and loading models
Classification models
- multilabel classification
- multiclass classification
Convolutional Neural Networks
- CNN theory
- develop an image classification model
- layer dimension calculation
- image transformations
- Audio Classification with torchaudio and spectrograms
Object Detection
- object detection theory
- develop an object detection model
- YOLO v7, YOLO v8
- Faster RCNN
Style Transfer
- Style transfer theory
- developing your own style transfer model
Pretrained Models and Transfer Learning
Recurrent Neural Networks
- Recurrent Neural Network theory
- developing LSTM models
Recommender Systems with Matrix Factorization
Autoencoders
Transformers
- Understand Transformers, including Vision Transformers (ViT)
- adapt ViT to a custom dataset
Generative Adversarial Networks
Semi-Supervised Learning
Natural Language Processing (NLP)
- Word Embeddings Introduction
- Word Embeddings with Neural Networks
- Developing a Sentiment Analysis Model based on One-Hot Encoding, and GloVe
- Application of Pre-Trained NLP models
Model Debugging
- Hooks
Model Deployment
- deployment strategies
- deployment to on-premise and cloud, specifically Google Cloud
Miscellanious Topics
- ChatGPT
- ResNet
- Extreme Learning Machine (ELM)