I published my notes for the [Guild] Artificial Intelligence (Fall 2020) lectures and the CS50’s Introduction to Artificial Intelligence with Python course.
My projects focus on planning because its techniques will be applied in my current CODE project Study Journey. I focused less on reasoning because there isn't much application of its techniques in our project.
-
Maze
- Inside the Planning/maze directory run
python maze.py maze[1-5].txt [DFS | BFS | HS]
- Inside the Planning/maze directory run
-
TicTacToe
- Inside the Planning/tictactoe directory run
python runner.py
to play against the ai.
- Inside the Planning/tictactoe directory run
- Crossword
- Inside the Optimization/minesweeper run
python generate.py data/structure1.txt data/words1.txt
, to specifying a structure file and a words file. If an assignment is possible, you should see the resulting assignment printed.
This project is an AI to create a crossword puzzle. It's technically speaking, not an optimization but a constraint satisfaction problem (CSP). It would be one if there were a continuous rating of a produced crossword puzzle. However, with this exercise, the outcome was binary.
- Minesweeper
- Inside the Reasoning/minesweeper directory, run
python runner.py
- Inside the Reasoning/minesweeper directory, run
-
Traffic
- Download the distribution code from https://cdn.cs50.net/ai/2020/x/projects/5/traffic.zip and unzip it.
- Download the data set for this project and unzip it. Move the resulting gtsrb directory inside of your traffic directory.
Inside the traffic directory, run
pip3 install -r requirements.txt
to install this project’s dependencies: opencv-python for image processing, scikit-learn for ML-related functions, and tensorflow for neural networks. - Inside of the Neural_Networks/traffic directory, run
python traffic.py gtsrb
to train the network.
This project includes a markdown file in which I document my experimentation process and describe how I was investigating different options.