My Projects

A short description of my projects

Homepage

I’m a Python and R developer interested in statistics —–> LinkedIn Profile

Demos

  1. Background Remover A docker container deployed on GCP Cloud Run (serverless) that can remove the background from an image.
  2. Excel demo A simple excel spreadsheet with slicers.
  3. Power BI Accounting Dashboard Also in pdf format
  4. JavaScript App A card matching game deployed on AWS with CI/CD pipeline.
  5. Natural Language Processing Demo A jupyter notebook utilizing Word2Vec to explore relationships between words in a twitter dataset.

HuggingFace Spaces

  1. Simple summary A HuggingFace Space that summarizes raw text.
  2. Wikipedia summary A HuggingFace Space that summarizes a Wikipedia article based on its title (titles without disambiguations work better).
  3. Pet classifier A HuggingFace Space that classifies images og dogs and cats.
  4. Image Captioning A HuggingFace Space that takes an image and produces a caption (relatively quickly).
  5. Text to Speech A HuggingFace Space that rapidly produces audio from text.

Current Projects

Vesuvius Challenge

The Vesuvius Challenge is a machine learning and computer vision competition that aims to extract passages of never-before-seen text from inside an unopened (and unopenable) carbonized scroll. The goal is to use computer vision and image segmentation techniques to extract faded ink from X-rays and other medical scans of papyri that were carbonized during the eruption of Mount Vesuvius.

Language Learning App

An app focused on grammar and vocabulary practice based on multiple choice questions (cloze). Starting with French and Spanish with translations provided in English. Currently working on a Flask implementation, which will then be ported into Flutter. Rough development draft follows.

News aggregator site

Currently setting up a news aggregator, to feature a rolling feed of breaking news in a 48-hour cycle.

Thesis

Here is a link to my thesis titled: “Modeling, Prediction and Categorization of elderly fall with deep learning techniques” (in Greek).

About page.