Machine Learning Research

Stocksmith is supported by:

An analysis of historic stock datasets and building a machine learning algorithm to predict stock behavior.

Project Structure

database-agg/

Scripts to add data to database with proper time and json formatting, has firebase scripts that will add data directly to firestore.

datasets/

The collection of datasets we are using for sample analysis.

r-analysis/

Multiple R files with a reference inital markdown and inital .docx files to generate reports as well as understand how to analyze large CSVs with R.

real-time-data/

Python API that uses Yahoo Finance API to get real time stock data. viz.py has plotting peaks of stocks over max 1 year. The folder also includes Go scripts for real time finance data scraping.

research-rnn/ Building and enchancing Recursive Neural Network models to predict stock prices.

Getting Started

These instructions will get you a copy of the project up and running on your local machine.

Prerequisites

• Python 3.6.9 or above with pip3
• Jupyter Notebook

Installing

For Mac OS/Linux users:

git clone https://github.com/stocksmith/ml-research.git

sudo bash setup.sh

For windows users:

https://github.com/stocksmith/ml-research.git

setup.bat

Built With

• Jupyter - Prototyping with Python