How joshwen7947/machine-learning-pipeline works
This is a small, single-script scikit-learn workflow for predicting vehicle CO₂ emissions. It reads a CSV, separates features and target, preprocesses numerical and categorical columns, trains and evaluates a random-forest regressor, saves the fitted pipeline, and displays evaluation plots. The stages are all implemented in one file; the edges reflect the calls and data flow visible there. The README gives no additional product capabilities, and the repository does not show how the script is launched.
Main entry points: pipeline.py Pipeline script · vehicle_emissions.csv Emissions CSV
Components 9
Input data
- Emissions CSV
ML_pipeline/vehicle_emissions.csv
Model workflow
- Pipeline script
ML_pipeline/pipeline.py - Feature preparation
ML_pipeline/pipeline.py - Preprocessing
ML_pipeline/pipeline.py - Model training
ML_pipeline/pipeline.py
Results
- Evaluation metrics
ML_pipeline/pipeline.py - Saved pipeline
ML_pipeline/pipeline.py - Evaluation plots
ML_pipeline/pipeline.py
Other components
- User
Connections 8
- User → Pipeline script: runs
- Pipeline script → Emissions CSV: reads
ML_pipeline/pipeline.py - Pipeline script → Feature preparation: separates
ML_pipeline/pipeline.py - Feature preparation → Preprocessing: configures
ML_pipeline/pipeline.py - Feature preparation → Model training: splits
ML_pipeline/pipeline.py - Model training → Evaluation metrics: predicts and scores
ML_pipeline/pipeline.py - Model training → Saved pipeline: saves
ML_pipeline/pipeline.py - Evaluation metrics → Evaluation plots: visualizes
ML_pipeline/pipeline.py