How ash7845/course5i works
This repository is an MLOps demonstration scaffold. Its visible workflow reads a configured CSV source, saves a raw dataset, then splits that dataset into training and test files. The README also describes feature engineering, model training and prediction, and visualization, but those modules are empty stubs in the excerpts, so their runtime connections are not asserted. The graph shows only evidenced calls and file operations; configured paths and external data sources are represented without assuming their actual values.
Main entry points: load_data.py Save raw data · split_data.py Split dataset · get_data.py Read source data · params.yaml Pipeline config
Components 13
Data preparation
- Pipeline config
params.yaml - Read source data
src/get_data.py - Save raw data
src/load_data.py - Split dataset
src/split_data.py - Raw dataset
- Train and test files
- Dataset processing stub
src/data/make_dataset.py
Modeling
- Feature engineering
src/features/build_features.py - Model training
src/models/train_model.py - Model prediction
src/models/predict_model.py
Visualization
- Visualize results
src/visualization/visualize.py
Other components
- Project user
- Configured CSV source
Connections 11
- Project user → Read source data: runs
src/get_data.py - Read source data → Pipeline config: reads
src/get_data.py - Read source data → Configured CSV source: loads CSV
src/get_data.py - Project user → Save raw data: runs
src/load_data.py - Save raw data → Read source data: calls
src/load_data.py - Save raw data → Pipeline config: reads
src/load_data.py - Save raw data → Raw dataset: writes
src/load_data.py - Project user → Split dataset: runs
src/split_data.py - Split dataset → Pipeline config: reads
src/split_data.py - Split dataset → Raw dataset: reads
src/split_data.py - Split dataset → Train and test files: writes
src/split_data.py