How rakshitanaik2006/mlops_telecom_customerchurnpredicition works
This repository contains a small churn-data workflow: the CLI loads a CSV and passes it to preprocessing, which removes duplicates, splits features and labels, transforms columns, and resamples the training set. The README describes a larger train/evaluate/predict pipeline, model registry, and REST/web serving, but the sampled model-stage modules are empty and the API path is absent from the file tree, so those capabilities are not represented as implemented components. The graph reflects only shown calls and documented dataset location; no persistence of preprocessing results is shown.
Main entry points: main.py CLI workflow · model_building.py Model building · data_ingestion.py CSV ingestion · churn - churn.csv Churn CSV
Components 8
Workflow entry
- CLI workflow
main.py
Churn data
- Churn CSV
data/raw/churn - churn.csv - CSV ingestion
src/data_ingestion.py
ML preparation
- Data preprocessing
src/data_preprocessing.py - Model building
src/model_building.py - Model evaluation
src/model_evaluation.py - Model prediction
src/model_prediction.py
Other components
- Pipeline operator
Connections 6
- Pipeline operator → CLI workflow: runs
README.md - CLI workflow → CSV ingestion: calls
main.py - CSV ingestion → Churn CSV: reads CSV
- CLI workflow → Data preprocessing: passes data
main.py - Model building → Model evaluation
- Model building → Model prediction