Skip to content

Latest commit

 

History

History
32 lines (25 loc) · 688 Bytes

README.md

File metadata and controls

32 lines (25 loc) · 688 Bytes
Raw

Timeseries Forecasting with Gaussian process regression

Objective: Demand Forecasting

Python packages:
Python-dateutil
Pytest
Isoweek
Sklearn
Pandas
Numpy
Datetime

Steps:

  1. Data processing
  2. Seasonality, Trend & Residue extraction
  3. Residue modelling
  4. Residue prediction
  5. Unit test and integration test results

Scripts information:

  • timeseries_modeling_analytics.py
    Arguments:
    sample.csv: sample data csv file
    n: required forecast period for step-2 and step-4
    Trigger command: python portcast_analytics.py sample.csv 6

  • test_cases.py
    Arguments: None
    Trigger command: pytest test_portcast.py