Skip to content

PaoloBenigni/deeplearning

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

53 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DeepCollege

License Binder Discord

Introduction

This project aims to create a collection of Jupyter notebooks discussing important topics in Deep Learning.

The tour sequence

  • Linear Classifier
  • Linear Regression
  • Support Vector Machine
  • Gradient Descent (No framework)
  • Forward props & Back props
  • Multilayer Perceptron
  • Neural Net (No framework, only numpy)
    • vectors & tensors
    • Different neural net layers (dense, dropout, max-pooling layers)
  • Data preprocessing (house price dataset, catvsdog, IMDB)
  • K Mean Clustering
  • CNN (Pytorch, TF, Keras) -> classify whales photos
  • Callbacks (early stopping and etc)
  • TF Dashboard
  • Training model in the cloud
  • Sentiment Analysis (binary classification)
  • Sentiment analysis (multi-classification)
  • Word2vec
  • Hyperparamter Tuning
  • Transfer Learning
  • Generate song lyrics and stories
  • Hyperparameter tuning
  • Feature engineering
  • Linear regression Kaggle competition using the above knowledge
  • Generative Models
  • GAN basic
  • StarGAN
  • CycleGAN
  • Reinforcement learning

Pre-requesties

  • Basic Python Knowledge
  • Some Machine Learning
  • Basic idea about Deep Learning

Installation

  1. Anaconda

Anaconda is a package management platform Data Scientists that lets you easily manage and install dependencies in cross-platform manner. It also ships with Jupyter Notebook, which plays a critical role in order to contribute to this project.

For more detail about why you should use Anaconda? https://www.quora.com/Why-should-I-use-anaconda-instead-of-traditional-Python-distributions-for-data-science

Installation per platform:

  1. Docker

Details will be added shortly

How do you contribute?

  1. Join the Discord channel https://discord.gg/MAMPnmm
  2. Goto #request-to-join channel and post your Github Account name!
  3. Once you are granted with access to the project, please create a git branch with your name
  4. Complete each topic and challenge in order of number sequence
    • for example: 000-Linear-Classification -> 001-Linear-Regression
  5. Reference existing code submissions from contributors or Wiki pages
  6. I will post Jupyter Notebooks with sample code or challenges to complete

About

deeplearning tour

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%