cs482-fall2018

Project Code For cs482/682 Fall 2018 @ UNR

project 1

Create a tic-tac-toe move-maker program. I've given you some skeleton file I/O code that you can use and the spec for some functions that you have to fill in. Make those functions work, and you should get an 'A'. Implements the minimax algorithm.

to compile

mkdir build && cd build
cmake ..
make

project 2

Create spam filter using niave Bayes classifier to give sms either ham or spam classification Project includes a training program that takes a .csv file of texts and parses it to train the Baye's classifier. Output are two files (one for spam and one for ham) with frequencies of words. The second file is the classifier that takes in a file as input and outputs a file that gives a classification of ham or spam for each sentence. The third file, addtotraining, takes in command line argument of a sentence to be added to the output files from the training program. Arguments: training -i -os -oh classify -i -is -ih addtotraining -is -ih -s ""

project 3

Implement a kalman filter to estimate state of 1D robot that has position x and velocity xdot. Sensor output is position only.

project 4

Implement Reinforcement learning, specifically Q-learning, using the OpenAI gym environment. First file cart.py is a solution to the cartpole balancing environment. Program learns how to balance pole on a cart for 200 ticks of time. Second file car.py is solution to mountain car. Program must learn how to get up a hill by rocking back and forth in a valley.