web-application-for-demand-forecasting

In this project in which I worked with Mehdi Hamdoune, we developed a web application to make commercial forecasts of a company from its history, using deep learning and machine learning algorithms. On our software, we made available to the user five forecasting algorithms while evaluating their accuracy and performance in order to obtain the best forecast's result.

• Front end: Angular, NodeJS.
• Back-end: Numpy, Pandas, Django, Tensorflow, Keras, Scikit-Learn, Matplotlib, etc.
• Forecasting methods : exponential smoothing (Single, Double, Holt-Winters) - LSTM - ARIMA.

Getting Started

Angular (FrontEnd) :

1 - Install Nodejs :

You can install NodeJs from the link below : https://nodejs.org/dist/v14.15.4/node-v14.15.4-x64.msi

2 - Run command line as administrator :

• Run command

npm install -g @angular/cli

• Go to the angular's file directory, for exemple :

d:

cd D:\Angular\Angular_FrontEnd

• Install the project (for only the first time) :

npm install

• Launch server :

ng serve -o

Python (Backend)

First of all, you have to check your python's librairies version, in case one is missing you have to install it with 'pip'.

The requierement versions :

-asgiref==3.3.1

-cycler==0.10.0

-Django==3.1.5

-django-cors-headers==3.6.0

-djangorestframework==3.12.2

-h5py==3.1.0

-joblib==1.0.0

-Keras==2.4.3

-kiwisolver==1.3.1

-matplotlib==3.3.3

-numpy==1.19.5

-pandas==1.2.0

-patsy==0.5.1

-Pillow==8.1.0

-pyparsing==2.4.7

-python-dateutil==2.8.1

-pytz==2020.5

-PyYAML==5.3.1

-scikit-learn==0.24.0

-scipy==1.6.0

-six==1.15.0

-sqlparse==0.4.1

-statsmodels==0.12.1

-threadpoolctl==2.1.0

Run command line as administrator

• Go to the python's file directory, for exemple :

d:

cd D:\Angular\pythonBackendGOP-main

• Launch server :

python manage.py runserver