Clothing Classification
A multi-category image classifier using a 3-layer Sequential Convolutional Neural Net.
Ryan Leyba
Data-Driven | Cognitive Science/Computational Modeling Background
I am a recent graduate from the University of California, Berkeley currently pivoting from an academic background in Cogntive Science with a concentration in Computational Modeling. I use my background to understand a dataset beyond its values; to understand the source of it. Being able to not only analyze the data itself but analyze the source allows me to bring unique insights in the EDA process to later provide novel findings down the pipeline.
Skills
Here are some of the skills I aquired in my coursework as an undergrad and as a Data Science Apprentice at General Assembly.
Projects
Here are some projects I worked on in the last couple months. Please click the image to be directed to the repo.
New Light Technologies Client Project
A web application that integrates multiple API's to extract data from a photo taken of a property that experienced a natural disaster and return the most recent photo of the property prior to the natural disaster.
Reddit Classification
A Sentiment Analysis and NLP project that analyzes the text of a reddit post to see if it came from either the "Venting" or "Confessions" subreddit
Ames, Iowa Housing Price Predictions
Participated in a "No Boosting" General Assembly Private Kaggle competition to predict housing prices. Placed 3rd out of 94 competitors
Contact Me
Feel free to reach out if you have any questions regarding my past projects