Skip to content

1337Farhan/CryptiWeb

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Django CI

CryptiWeb

The web app and enhanced version of the Crypti project.

1. Introduction 👋

Developed by a team consisting of Me, @Anas-Elhounsri, and @mosman4, CryptiWeb is an online version of my previous project Crypti.

2. Setup

Note: you need to have our AWS setup ready before doing this, refer to Crypti-LSTM

There are four steps to host CryptiWeb on your local machine.

  • Create a conda environment with requirements.txt (ipython is recommended)

    conda env create -f requirements.txt python=3.11 ipython

  • Create a secret key from djecrety and replace the SECRET_KEY in settings.py

  • Setup environment variables in your terminal

    export AWS_ACCESS_KEY_ID="<your_s3_access_key_id>";

    export AWS_SECRET_ACCESS_KEY="<your_s3_secret_key>";

    export AWS_DEFAULT_REGION="your_aws_region";

  • Run update_coins() from tasks.py in your shell to fetch prediction from aws

    $ python3.11 manage.py shell

    >>> from coin_prediction.tasks import update_coins()

    >>> update_coins()

Finally, you can run python3.11 manage.py runserver.

3. Differences from - the OG - Crpyti 📈

CryptiWeb is a web app (This repo) developed using The Django framework to allow more users to get to know and try the project. It realizes a Long Short Term Memory (LSTM) model Crypti-LSTM to perform time series prediction instead of linear regression, developed and hosted on Amazon Web Services (AWS) with an architecture designed by @Anas-Elhounsri. And finally, the project is available as a mobile app CryptiMobile designed by @mosman4.

4. Known issues 🐛

Although we have moved from using linear regression to using LSTM, we know that our model is inaccurate because it only consumes one variable (Price). In the near future, we plan on updating our model to a Multivariate LSTM including more variables for more accurate forecasting results.

Read more:

5. Roadmap 🚀

We are planning to achieve the following by the end of '23

  • Update the model to a Multivariate LSTM
  • Add more information to the website
  • Enhance the look and feel of the app
  • Conduct testing and re-adjustments on the model

6. Disclaimer ⚠

This project is for educational, entertainment, and testing purposes only, it is by no means a financial advice/tool and should not be relied on for any financial decisions.

Finally, we will be more than happy to see your contributions.

About

Online version of the Crypti project

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors