an application that predicts the share price in the future
This application is targeted for the JSE Market
.
this application wont run in Windows properly due to the MVC Compiler for C++
but you can work around that by using Anaconda. Not a fan of it and i wont be using it so in development i used Ubuntu Linux
so follow the steps in usage and you can get this app working. This app also runs in AWS.
Take a look at the notebooks etc to see how i planned my app. I tried many routes i found Prophet
was the fastest in terms of accuracy and speed.
The Dash app is imcomplete - i had plans to merge the ADGSDK into Dash. Orignal idea was to use Keras
and Dash
as the Data Science Stack.
If you want to this application to predict far greater years of forecasting, git clone
my repo.
To run app once cloned
you need to install requirements you can do this by running pip3 -r requirements.txt
Running app
streamlit run streamlitapp.py
- For
predicting-stockprice-neuralnetwork.ipynb
- For main web app
streamlitapp.py