MNK/portfolio
Machine learning

Freight Rates Predictor

End-to-end scikit-learn pipeline

End-to-end ML pipeline on spatial, load, market-index and temporal features.

Code

Problem

Predict freight rates from load and market data.

Approach

  1. Features built from spatial, load, market-index and temporal data
  2. End-to-end scikit-learn pipeline from raw data to prediction
Freight Rates Predictor workflow: 2 stages from input to result
Workflow

Results

  • Complete training-to-prediction pipeline

Stack

  • Python
  • scikit-learn
  • Feature engineering

By Muzammil Nawaz Khan, AI Engineer.