Skip to content

eeshawn11/eeshawn11

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

38 Commits
 
 
 
 
 
 

Repository files navigation

Hi there 👋

I am a Data Science Immersive graduate from General Assembly Singapore and new to the data community! Coming from a Hospitality background, this has been quite a major shift in my career trajectory, but I've been fully enjoying my time picking up new skills like Python and SQL as I navigate my way through these uncharted waters.

LinkedIn Logo Happy to connect on LinkedIn. 😄

📸 Fun fact: I enjoy photography as a hobby, especially while travelling. Check out some of my photos here and there.

Top Langs


📈 Data Science Projects

Exploration of data retrieved from Data.gov.sg, in particular HDB resale prices from 2012, to create an interactive dashboard. I am also hopeful that any insights gleaned could be helpful to my own HDB purchase journey, but no fancy prediction model here (yet?), just some charts and maps.

This is an ongoing project to document my learning with using Streamlit and various Python libraries. While such a dashboard could perhaps be more easily created using PowerBI or Tableau, I am also taking the opportunity to explore the various Python plotting libraries and understand their documentation.

Streamlit App

Skills Demonstrated:

  • Data extraction through live Data.gov.sg API
  • Data transformation with pandas
  • Working with geospatial data using OneMap API, Shapely and Mapbox
  • Data visualisation with Vega-Altair and Plotly
  • Web app deployment with Streamlit

Time series forecast using a dataset gathered from the Energy Market Authority of Singapore. The original data is stored in individual weekly Excel files by month and year, which I have scraped from the website and combined into a dataset.

This is a relatively straightforward univariate time series problem involving electricity demand in Singapore and Prophet achieved a Mean Absolute Percentage Error (MAPE) of 3.67% when predicting for demand in January 2023.

Kaggle

Skills Demonstrated:

  • Extracting data through Web scraping with Selenium
  • Data transformation with pandas
  • Loading and storing data into Kaggle dataset
  • Time series forecast with Facebook Prophet

Training and deploying a machine learning model that can recognise 12 basic hand seals from the Naruto anime using transfer learning with YOLOv8.

Open in Spaces

Skills Demonstrated:

  • Collecting and building my own dataset using OpenCV
  • Image annotation with LabelImg
  • Transfer learning using a pre-trained YOLOv8 model

A collection of various projects completed during my 12-week immersive with General Assembly.

Skills Demonstrated:

  • Supervised machine learning with scikit-learn, with a variety of regression, classification and ensemble models
  • Natural Language Processing (NLP) techniques with NLTK
  • Presentation of findings and recommendations at the end of each project
  • Collaboration and version control with git

Capstone project from Data Science Immersive. Trained an image caption generator with attention mechanism that achieved a BLEU-1 score of 0.52 on the Flickr30k dataset.

Streamlit App

Skills Demonstrated:

  • Deep learning, computer vision and natural language processing with TensorFlow and Keras
  • Handling of large dataset using TensorFlow's tf.data API
  • Deployment of custom Keras model to Streamlit web app

📈 Side Projects

Developed a game to familiarise with Object-Oriented Programming (OOP) and working with Classes. While the game concept was simple, it was interesting figuring out how to code various aspects of the game, such as collision detection and handling user input.

Skills Demonstrated:

  • Designing and coding game mechanics with PyGame, such as collision detection
  • Object-Oriented Programming in Python
  • Program testing and debugging

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published