NMA_Deep_Learning_Workshop

---

description: No description available. — — toc: false —

«««< HEAD

Neuromatch Academy Deep Learning Workshop - Summer 2023

This repository contains the exercise notebooks from the Neuromatch Academy Deep Learning workshop I attended in Summer 2023. The workshop covered a range of topics in deep learning, from the basics of PyTorch to more advanced concepts like transformers and reinforcement learning.

Workshop Materials

The materials are organized by week and day. Each day typically includes multiple tutorials and sometimes bonus tutorials.

Week 1: Basics and Linear Deep Learning

• Day 1: Basics and PyTorch
  • W1/D1/W1D1_Tutorial1.ipynb: PyTorch
• Day 2: Linear Deep Learning
  • W1/D2/W1D2_Tutorial1.ipynb: Gradient Descent and AutoGrad
  • W1/D2/W1D2_Tutorial2.ipynb: Learning Hyperparameters
  • W1/D2/W1D2_Tutorial3.ipynb: Deep linear neural networks
• Day 3: Multi Layer Perceptrons
  • W1/D3/W1D3_Tutorial1.ipynb: Biological vs. Artificial Neural Networks
  • W1/D3/W1D3_Tutorial2.ipynb: Deep MLPs
• Day 5: Optimization
  • W1/D5/W1D5_Tutorial1.ipynb: Optimization techniques

Week 2: Regularization, Convnets, Generative Models and Transformers

• Day 1: Regularization
  • W2/D1/W2D1_Tutorial1.ipynb: Regularization techniques part 1
  • W2/D1/W2D1_Tutorial2.ipynb: Regularization techniques part 2
• Day 2: Convnets and DL Thinking
  • W2/D2/W2D2_Tutorial2.ipynb: Deep Learning Thinking 1: Cost Functions
• Day 3: Modern Convnets
  • W2/D3/W2D3_Tutorial1.ipynb: Learn how to use modern convnets
• Day 4: Generative Models
  • W2/D4/W2D4_Tutorial1.ipynb: Variational Autoencoders (VAEs)
  • W2/D4/W2D4_Tutorial3.ipynb: Image, Conditional Diffusion and Beyond
• Day 5: Attention and Transformers
  • W2/D5/W2D5_Tutorial1.ipynb: Learn how to work with Transformers
  • W2/D5/W2D5_Tutorial2.ipynb: Understanding Pre-training, Fine-tuning and Robustness of Transformers

Week 3: Time Series, NLP, Unsupervised Learning and Reinforcement Learning

• Day 1: Time Series and Natural Language Processing
  • W3/W3D1_Tutorial1.ipynb: Introduction to processing time series
  • W3/W3D1_Tutorial2.ipynb: Natural Language Processing and LLMs
  • W3/W3D1_Tutorial3.ipynb: Multilingual Embeddings
• Day 2: DL Thinking 2
  • W3/W3D2_Tutorial1.ipynb: Deep Learning Thinking 2: Architectures and Multimodal DL thinking
• Day 3: Unsupervised and self-supervised learning
  • W3/W3D3_Tutorial1.ipynb: Un/Self-supervised learning methods
• Day 4: Basic Reinforcement Learning
  • W3/W3D4_Tutorial1.ipynb: Basic Reinforcement Learning
• Day 5: Reinforcement Learning for Games & DL Thinking 3
  • W3/W3D5_Tutorial1.ipynb: Reinforcement Learning For Games
  • W3/W3D5_Tutorial2.ipynb: Deep Learning Thinking 3
  • W3/W3D5_Tutorial3.ipynb: Planning with Monte Carlo Tree Search

Bonus Tutorials

• Bonus_Tutorial1.ipynb: Deploying Neural Networks on the Web

Datasets

• afhq/: Contains the “Animals Faces HQ” dataset, split into training and validation sets for cats, dogs, and wild animals.
• data/cifar-10-batches-py/: Contains the CIFAR-10 dataset in Python format.
• FashionMNIST/raw/: Contains the FashionMNIST dataset.

Additional Materials

• Models and Data sets — Neuromatch Academy_ Deep Learning.pdf: A PDF document related to models and datasets used in the workshop.

  Neuromatch_Academy_Deeplearning_workshop

  eb2ec683c55123c5e0ee55a184cc40d1f272e801