Skip to content

Latest commit

 

History

History
68 lines (44 loc) · 4.08 KB

README.md

File metadata and controls

68 lines (44 loc) · 4.08 KB
Raw

First Machine Learning Project

Introduction

My first three years of College and majoring in Computer Science I had one goal of expanding my breadth of knowledge in the Computer Science realm.
I have explored the fields of Scientific Computing, Android Application Development, Assembly and Computer System principles, Game Design and Visual Programming, web design, Encryption, etc. I have gained an (at least basic) intuitive understanding of each of these fields to develop myself as a well rounded Computer Scientist. For my senior year, I am focusing on further exploring depths of certain fields and make myself ready as a competent competitor for namely two categories : IOS/Android Application development and Artifical Intelligence.

For the Summer of 2020 I decided to keep myself very busy with my fields of interests. In the prior 2/3rds of the Summer, I explored fields such as Graph Theory and Finite Calculus, built my own website, and completed a multidude of Computer Science Interview Preperation Questions.

During the first 2/3rds of Summer 2020 I also:

  • Impletemented various algorithms from scratch (in C++ or Java) such as:

    • Djikstra's algorithm
    • BFS & DFS
    • Quicksort and Mergesort

  • Read the works of Great Philosophers such as:

    • Immanuel Kant
    • René Descartes
    • Aristotle
    • Confucius

  • Changed my morning routine to include Meditation and Cold Showers

  • Stayed Physically active and exercised most days of the week

  • Made time to have fun too!

The Final Third of Summer 2020 I plan to implement my knowledge and buld projects to feature on my resume.

  1. Machine Learning Stock Predictor
  2. Social Voting Theory Android Application
  3. Philosophical Essay

Summer 2020 Project 1 : Machine Learning Stock Predictor

I have designed a 'course' to learn the fundamentals of Machine Learning:

  1. Complete Google Machine Learning Crash Course

  2. Simple/fun ML project to have hands on experience

  3. Andrew Trask's Grokking Deep Learning

  4. More sophisticated ML project

    option 1
    5. Pedro Domingo college course for Machine Learning (found on Youtube)
    6. Finish all HW assignments of this course

    option 2
    5. Aurélien Géron Hand's On Machine Learning edition 2
    6. Complete plenty of exercises in this book.

This project is the second on the list: A simple/fun ML project to have hands on experience

The use of Artifical Intelligence for Stock Market Analysis/Prediction is a very big and competitive field. There are resources and websites which act as communities and API for developing your own algorithms and papers (such as quandl and quantopian) however my goal for this project is not to develop a industry standard algorithm, but rather have hands on experience with Machine Learning and the tools involved

I will be using TensorFlow, Pandas, Numpy, and Matplotlib for the development and analysis of the algorithm, as well as BeautifulSoup to scrape valuable information from stock analysis websites (such as finviz). Big thanks to Nicolas P. Rougier for 100 Numpy Exercises and Alex Riley for 100 Pandas Exercises which acted as tutorials for using Numpy and Pandas correctly for Machine Learning Projects.

Information on my developed algorithm, such as features, representation, evaluation, failure scenarios, ideal outcomes, bias prevention, etc can be found on the Jupyter Notebook on this repository.

Hope you enjoy!

Note: As I was working on both a Java project and a Python project at the same time, I did my best to adhere to code conventions of the individual projects. However, there is a possibility I camel-cased something in Python, or I snake-cased something in Java and I was unable to catch it doing read throughs of my code due to working on both projects at the same time. If there are conventions I am not following please feel free to contact me!

Contact: [email protected]