This project implements an end-to-end MLOps pipeline for engine predictive maintenance: classify engine condition (Normal vs Maintenance Required) from sensor readings (RPM, lubricating oil pressure, fuel pressure, coolant pressure, oil and coolant temperatures). The pipeline covers data registration, exploratory data analysis, data preparation, model building with experimentation tracking (MLflow), model deployment (Streamlit on Hugging Face Spaces), and automated workflows (GitHub Actions).
| Path | Description |
|---|---|
Engine_PM_Interim_Notebook_final.ipynb |
Main notebook: EDA, prep, training, deployment, and report sections |
engine_pm_project/data/ |
Raw data folder (e.g. engine_data.csv) |
engine_pm_project/model_building/ |
Scripts: data_register.py, prep.py, train.py |
engine_pm_project/deployment/ |
Dockerfile, Streamlit app.py, requirements.txt, deploy_to_hf_spaces.py |
.github/workflows/pipeline.yml |
CI: register-dataset → data-prep → model-training → deploy-hosting |
| Resource | Link |
|---|---|
| GitHub repository | github.com/ananttripathi/engine-pm-project |
| Hugging Face Space (Streamlit app) | huggingface.co/spaces/ananttripathiak/engine-pm-streamlit |
| Hugging Face dataset | huggingface.co/datasets/ananttripathiak/engine-pm-data |
| Hugging Face model | huggingface.co/ananttripathiak/engine-pm-model |
cd /path/to/Interim_Submission
jupyter nbconvert --to notebook --execute --inplace Engine_PM_Interim_Notebook_final.ipynbPlace title_page.png in the same directory as the notebook so the image displays correctly.
- Streamlit app: The Space runs the app in
engine_pm_project/deployment/(Docker). It loads the model from the Hugging Face model hub and predicts from six sensor inputs. - CI: Push to
mainruns the pipeline; set HF_TOKEN in GitHub Secrets for data/model/Space uploads.
See repository for license information.