CodeAcademy AI course material

This course cover an introduction to Artificial Intelligence - from Python Crash Course to Generative Deep Learning.

Index

List of all covered topics

Intro to Python
Introduction to syntax
Lists, Sets and Tuples
Functions
Generators
Classes
Classes II
Classes III
Classes IV
Classes V
SQL with Python
Scraping

Intro to Numpy
Introduction to Numpy
Numpy II

Intro to Machine Learning
Linear Regression
Wine Quality Prediction with Linear Regression
Wine Quality Prediction with Linear Regression II
Titanic Survivors Prediction
Sklearn Pipelines and Transformers
Housing price prediction
IMDB Movie Review Sentiment Analysis
Decision Trees and Random Forrest Classifier
Decision Trees and Random Forrests
AdaBoost and GradientBoost
XGBoost

Intro to Deep Learning
Neural Nets from scratch
Regularization I
Building Micrograd
Building Micrograd II
MLP
Initialization
Text generation with Bigrams
Text generation with Bigrams II
Pytorch Lightning
Weight initialization and batch norm
PyTorch modules
Manual Backprop
MLP II
Transfer Learning
Transfer Learning II
Huggingface demo
Images classification with Huggingface
Image Captioning
Text summarization

Related repositories

Title	Description	Link
Simple Python Project	Project that showcases Python project structure and usability of modules	https://github.com/dqmis/simple_python_project
Dog Shelter	Repository to showcase FastAPI functionality	https://github.com/dqmis/dog_shelter
Recipes Scraper	Project showcasing web scraping with Python	https://github.com/dqmis/recipes_scraper
Recipes Classifier Service	Project showcasing FastAPI and Sklearn integration	https://github.com/dqmis/recipes-classifier
Text classification with Transformers	Project showcasing text classification with Transformers and comparing it to classical ML approaches	https://github.com/dqmis/nlp-classification
Text generation with Mini-GPT	Project showcasing text generation with Mini-GPT built from scratch	https://github.com/dqmis/gpt-demo
Deployment to GCP demo	Demo application for building and deploying simple python app to GCP Cloud Run	https://github.com/dqmis/fastapi-gcp-demo

Set up

Install VSCode

• Follow instructions presented in the website.
• Install Python, Jupyter, autoDocstring, Mypy, Pylance extensions

Set up Git

Follow instructions presented in the website.

Set up Python

• Install Python (if you use Windows)
  • Go to the official webpage and download Python for windows.
  • Follow the instalation process. IMPORTANT - Check Add Python to PATH checkmark
• Set up Pyenv:
  • OSX and Linux installation:
    • Follow the installation guide from here: https://github.com/pyenv/pyenv#installation
  • Windows installation:
    • Follow the installation guide from here: https://github.com/pyenv-win/pyenv-win
  • $ pyenv install 3.9.13
  • If you want to set the global Python version - $ pyenv global 3.9.13
• (Not recommended) Alternative - install a specific version of Python from https://www.python.org/downloads/

Set up Poetry

• Install (full guide: https://python-poetry.org/docs/#installation)
• Configure:
  • $ poetry config virtualenvs.in-project true - creates the virtual environment in the current folder as opposed to somewhere random
  • $ poetry self add poetry-dotenv-plugin - enabled sourcing environment variables from .env when commands run through Poetry
  • $ poetry config virtualenvs.prefer-active-python true seems to make Poetry work better with Visual Studio Code

Install dependencies

• Go to this project's dir with your terminal: $ cd <PATH_TO_THE_PROJECT>
• $ poetry install (this will create a poetry.lock file and a virtual environment in the .venv folder)
• If you use OSX / Linux
  • $ source .venv/bin/activate this will activate a newly created environment
• If you use windows
  • .venv\Scripts\activate this will activate a newly created environment
• Alternatively run $ poetry shell to initialize the environment.

Set-up pre-commit

• $ pre-commit install

Start Jupyter Lab and run notebook

• Start Jupyter Lab: $ jupyter lab
• Execute Notebook inside notebooks dir