Harrison Gropper's Projects
Intro to AI Assignment, unknown blocked cells/robot movements. This was a part of a class project where we were given a specific scenario and asked to compute the corresponding probabilities.
At the start of the coronavirus many people were scrambling to gather computer parts. As a result the prices of GPU's (Graphics Processing Units) sky rocketed. This high demand lead to high prices, and people buying them up like crazy. Note that these GPU's went out of stock right away... within seconds! Could I make a bot to buy one for myself?
Implementing Decision Trees and Regression from scratch. Decision Trees Question: There are multiple features... which feature do we split on? AND where do we split in this feature?
The goal of this project was to find a correlation between stocks and their current popularity using google trends.
Config files for my GitHub profile.
Filling holes in images with missing parts. These missing parts are random blank spaces in the image. Catch: Everything has to be coded yourself (from scratch)! No external machine learning libraries!
At all times students have a certain number of remaining sessions left. Mathnasium does not have an automatic system to let the parents know that their kids are running low on sessions. How could we easily pull the data from the scheduling website to access this data, so then we can notify the parents ourselves??
A terminal based GUI that uses a headless version selenium to send text messages to parents. The message is usually about a student being late/not showing up.
During COVID where I work (Mathnasium) a math learning center became half online and half in person. I knew that the sudden transition online would contribute a lot of stress and extra work for the team. I took it upon myself to automate a very tedious task to save tons of time!
This was a homework assignment in my machine learning class. The homework included the topics of information gain, soft classifiers, gradient descent, and support vector machines.
We look at the structure of data and not its corresponding output Y. By looking at the structure of data we can better understand the given data. We can do this using various unsupervised machine learning algorithms, but in this case I am using an autoencoder to model the structure of the inputs.