Deep-Learning-Material

This repository contains jupyter notebooks and other materials taught in TA sessions of Deep Learning course (CE719) instructed by Dr. Beigy at Sharif University of Technology.

Session 1

• Introduction to deep learning frameworks
• Introduction to Python & NumPy
• Introduction to PyTorch
• video link : https://drive.google.com/file/d/1iYMNg3iiCZ45lAxildljWq6ZTuPq8FZu/view?usp=sharing

Session 2

• PyTorch modules
• An example of MLP
• video link : https://drive.google.com/file/d/1eZUI3hEqXAsUdSo9lll-8OgjyoquEoB6/view?usp=sharing

Session 3

• HW1 description
• Optimizers description
• Video and files link: https://drive.google.com/file/d/1rdHf7lrtoxSCxlAoRlEi6ofhMwez7wA5/view?usp=sharing

Session 4

• hardware requirements for NN
• introduction to CNN with pytorch
• Video and files link: https://drive.google.com/file/d/1kRSUW4zBi5o2-37QoYeQf1B4dB6konT_/view?usp=sharing

Session 5

• Types of Convolutional Neural Nets
• Custom Dataloader, Transforms
• Normalization
• Transfer Learning
• train and test SVHN dataset ----> pkg version : torch==1.1.0 , tensorflow-gpu==1.10.0
• Video link (part1): https://drive.google.com/file/d/17VbrdsSOg-YRC38cO2YXG3NZ_QWu-Al8/view?usp=sharing
• Video link (part2): https://drive.google.com/file/d/1lnYPCS9bBznIPCvpwJdSUAsMuUapJfv6/view?usp=sharing

Session 6

• HW2 description

Session 7

• Intro to Recurrent Neural Networks
• Intro to TorchText
• Sentiment Analysis with LSTM
• Custom Loss function
• Video link : https://drive.google.com/file/d/11SVYuo5AwhwvRUeY-avUeR3CUXYprIlb/view?usp=sharing

Session 8

Optional Material

• Advanced topics in RNN
• Video Link (part1): https://drive.google.com/file/d/1jUf88cZBsItppDlSyeV1bRL01ICj4574/view?usp=drivesdk
• Video Link (part2): https://drive.google.com/file/d/10RcDVK7La5k-0guxilO5-bjeFa2lnRlN/view?usp=drivesdk

Necessary Material

• Implementation of Neural Machine Translation
• Tutorial Link: https://pytorch.org/tutorials/intermediate/seq2seq_translation_tutorial.html

Session 9

• Neural Networks & Debugging
• Video Link: https://drive.google.com/file/d/1E0fAigYlL1brF9NK7rq4mzIwGayT0oM7/view?usp=sharing

Session 10

• Intro to Graph Neural Nets
• Intro to NetworkX, Pytorch Geometric
• A Simple GCN for node classification
• Video Link: https://drive.google.com/file/d/1EY7ybU1EvvxMYJLx3OJykmWKzIpiDdP7/view?usp=sharing

Session 11

• Variational AutoEncoder (VAE)
• Theory of VAE
• pytorch code for VAE on MNIST dataset
• Video Link: https://drive.google.com/file/d/1NDiaGuun6z7L7VikWfs71hmpNN0PUmo6/view?usp=sharing

Session 12

• VAE and GAN
• Theory of VAE and GAN
• Be Sure to Read "Correction.pdf" !
• BigGan Pre-trained on IMAGENET Demo.
• video Link: https://drive.google.com/file/d/1AWNwDeLo-2ZOsyPcyU4NhRpTbIHr6V_m/view?usp=sharing