=======

Demand forecasting with python

Develop a software that allows you to :

• Make commercial forecasts from a history
• Compare several forecasting methods
• Display the results (forecasts and comparison)

Forecasting algorithm(s) used:

• Single, double and triple exponential smoothing with parameter optimisation
• ARIMA to compute a statistic model
• LSTM, a deep learning model

Features

• Give the decision-maker the choice of which algorithm(s) to use
• Propose default settings, and give the decision maker the decision maker to change the parameters

Requirements

• Pillow==8.1.0
• matplotlib==3.3.3
• pandas==1.2.0
• PyQt5==5.15.2
• PyQt5-stubs==5.14.2.2
• pyqt5-tools==5.15.2.3
• pyqtgraph==0.11.1
• scipy==1.6.0
• sklearn==0.0
• statsmodels==0.12.1
• tensorflow-keras==1.12.0
• XlsxWriter==1.3.7

python version for this project: 3.7.4

=======