LFD473: PyTorch in Practice - An Applications-First Approach

Part I: Training a Model in PyTorch

• PyTorch, Datasets. and Models

• Building Your First Dataset

  • Lab 1A: Non-Linear Regression / Solution

• Training Your First Model

  • Lab 1B: Non-Linear Regression / Solution

• Building Your First Datapipe

  • Lab 2: Price Prediction / Solution

Part II: Transfer Learning

• Transfer Learning and Pretrained Models

  • Lab 3: Classifying Images / Solutions

• Pretrained Models for Computer Vision

• Pretrained Models for Natural Language Processing

• Lab 4: Sentiment Analysis / Solutions

Part III: Computer Vision

• Image Classification with Torchvision

• Fine-Tuning Pretrained Models for Computer Vision

• Serving Models with TorchServe

• Datasets and Transformations for Object Detection and Image Segmentation

  • Lab 5A: Fine-Tuning Object Detection Models / Solutions

• Models for Object Detection and Image Segmentation

  • Lab 5B: Fine-Tuning Object Detection Models / Solutions

• Models for Object Detection Evaluation

Part IV: Natural Language Processing

• Word Embeddings and Text Classification

  • Lab 6: Text Classification / Solutions

• Contextual Word Embeddings with Transformers

• HuggingFace Pipelines for NLP Tasks

  • Lab 7: Document Q&A / Solutions

• Question and Answer, Summarization, and LLMs