Skip to content

Flask/React machine learning stock predictor web app.

Notifications You must be signed in to change notification settings

Jaspvr/Stock-Predictor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

56 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Machine Learning Stock Predictor Web Application

Demo and Code Walk-Through:

https://youtu.be/IHdjdF6lIEg

Overview

Flask web app with Python and Sklearn backend model that uses metrics of the specified stock, different sectors, and overall market over the past 20 years to predict whether a stock will increase over the next day and week. The Frontend is built in React and allows a user to input a stock ticker, and have the predictions output to the screen. The precision of these predictions is also output to the screen as it is found through backtesting when the user inputs the stock.

To Run it

Step 1: Create a virtual environment using the following commands in the terminal: Mac: python3 -m venv myenv Windows: python -m venv myenv

Step 2: Install all dependencies listed in dependencies.txt (flask, flask-cors, scikit-learn, pandas, yfinance, matplotlib): pip install package-name

Step 3: Run the backend flask application (run App.py). You should see the message: Running on http://127.0.0.1:5000

Step 4: Navigate to the react app and run it: cd react-stock npm start It should now be available at http://localhost:3000/

Check list

  • Find somewhere to scrape access data from
  • Determine which stocks to take in based on market cap and long term stability
  • Decide which metrics to take in for each stock
  • Create simple html file and connect to python with flask
  • Use machine learning to make stock price predictions
  • Back Test the algorithm
  • Add in sectors going up or down over the same time period ( and see if precision improves )
  • Add in other sector info that improves precision
  • Tech sector
  • Energy sector
  • Financials sector
  • Industrial sector
  • Real Estate sector
  • Create simple html frontend to verify flask app is working as expected
  • Create simple React front end
  • Connect React to Python
  • Send all necessary information from Flask to React frontend. App is working as expected at this point
  • Make the frontend presentable
  • Host the application
  • Add error handling message
  • Add Video Demo

About

Flask/React machine learning stock predictor web app.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published