mlops-project-generator 1.0.8

A CLI tool that generates production-ready MLOps project templates for Scikit-learn, PyTorch, and TensorFlow

🧠 MLOps Project Generator

Generate production-ready MLOps project templates — via Web UI or CLI.
Scikit-learn · PyTorch · TensorFlow · Multi-cloud · Zero setup required.

This stack supports the full MLOps lifecycle:

Data → Train → Track → Orchestrate → Deploy → Monitor → Improve

---

🚀 Features

• 🔧 Framework Support: Scikit-learn, PyTorch, TensorFlow/Keras
• 📊 Task Types: Classification, Regression, Time-Series, NLP, Computer Vision
• 🔬 Experiment Tracking: MLflow, W&B, Custom solutions
• 🎯 Orchestration: Airflow, Kubeflow
• 🚀 Deployment: FastAPI, Docker, Kubernetes, Cloud platforms
• 📈 Monitoring: Evidently AI, Custom solutions
• ☁️ Cloud Deployment: Multi-cloud templates (AWS, GCP, Azure)
• 🌐 Web UI: Browser-based generator — no install needed
• 🤖 CI/CD Ready: Non-interactive CLI mode for DevOps pipelines
• ⚙️ Config Presets: Save, load, and reuse project configurations
• 🎨 Custom Templates: Create and manage framework templates

---

� MLOps Presets

Choose from 6 pre-configured MLOps stacks tailored for different use cases:

Preset	Framework	Tracking	Orchestration	Deploy	Monitor
⚡ Quick Start	Sklearn	None	None	FastAPI	None
🧪 Data Science	Sklearn	MLflow	None	FastAPI	Custom
🧠 Deep Learning	PyTorch	W&B	None	Docker	None
📡 Production MLOps	PyTorch	MLflow	Airflow	Docker	Evidently
🏢 Enterprise	TensorFlow	MLflow	Kubeflow	Kubernetes	Evidently
🔬 Research	PyTorch	W&B	None	FastAPI	None

Use presets via CLI: mlops-project-generator init --preset <name> or select them in the Web UI.

---

�🌐 Web UI (Recommended)

The fastest way to generate a project — open it in your browser, fill in the form, and download a ready-to-use ZIP.

Run locally

cd web_ui
npm install
npm run dev
# → http://localhost:3000

Self-host / Deploy

The web UI is a Next.js 15 app with no external backend dependency. The generator runs entirely as Next.js API routes — deploy to Vercel, Netlify, Railway, or any Node.js host:

cd web_ui
npm run build
npm start

Vercel one-click deploy: set the root directory to web_ui and it works out of the box.

Web UI Tech Stack

Layer	Technology
Framework	Next.js 15 (App Router)
Styling	Tailwind CSS v4 + Glassmorphism
Forms	react-hook-form + zod
Generator	Pure TypeScript (no Python runtime needed)
ZIP output	archiver

---

📦 CLI Installation

Prefer the command line? Install the Python package:

From PyPI

pip install mlops-project-generator

From Source

git clone https://github.com/NotHarshhaa/MLOps-Project-Generator.git
cd MLOps-Project-Generator
pip install -e .

Development Install

pip install -e ".[dev]"

---

🎯 CLI Quick Start

Interactive mode (recommended for beginners)

mlops-project-generator init

Non-interactive mode (CI/CD)

mlops-project-generator init \
  --framework pytorch \
  --task-type classification \
  --tracking mlflow \
  --orchestration airflow \
  --deployment docker \
  --monitoring evidently \
  --project-name enterprise-ml \
  --author-name "ML Team" \
  --description "Production ML pipeline"

Available CLI Commands

Core

Command	Description
init	Generate a new MLOps project
validate	Validate an existing project structure
version	Show version

Configuration Management

Command	Description
save-preset <name>	Save current config as a preset
list-presets	List all presets
load-preset <name>	Load a preset
delete-preset <name>	Delete a preset

Template Management

Command	Description
create-template <name> <framework>	Create a custom template
list-templates	List custom templates
delete-template <name>	Delete a template
add-template-file <template> <path>	Add a file to a template

Analytics & Cloud

Command	Description
stats	Show generation statistics
analyze <path>	Analyze a generated project
cloud-services	List available cloud services
cloud-deploy <provider> <service>	Generate cloud deployment templates

init Flag Reference

Flag	Short	Values
--framework	-f	sklearn, pytorch, tensorflow
--task-type	-t	classification, regression, timeseries, nlp, computer-vision
--tracking	-r	mlflow, wandb, custom, none
--orchestration	-o	airflow, kubeflow, none
--deployment	-d	fastapi, docker, kubernetes
--monitoring	-m	evidently, custom, none
--project-name	-p	Any valid identifier
--author-name	-a	Any string
--description	--desc	Any string

---

🖼️ Screenshots

CLI Commands

Scikit-learn Generation

PyTorch Generation

TensorFlow Generation

---

📁 Generated Project Structure

your-project/
├── src/
│   ├── data/              # Data loading utilities
│   ├── models/            # Model implementations
│   ├── features/          # Feature engineering (sklearn)
│   ├── utils/             # Training utilities (pytorch/tensorflow)
│   ├── train.py           # Training entry point
│   └── inference.py       # Inference / FastAPI server
├── configs/
│   └── config.yaml        # Project configuration
├── data/
│   ├── raw/
│   ├── processed/
│   └── external/
├── models/
│   ├── checkpoints/
│   └── production/
├── notebooks/
├── tests/
│   └── test_model.py
├── cloud/                 # Cloud deployment files (if selected)
│   └── <provider>/<service>/
│       ├── Dockerfile
│       ├── cloud-config.yaml
│       └── deploy.sh
├── requirements.txt
├── Makefile
├── .gitignore
├── .env.example
├── project_config.json
└── README.md

---

🛠️ Framework-Specific Features

Scikit-learn

• RandomForest / Ridge with cross-validation and joblib serialization
• DataLoader + FeatureEngineer classes ready to extend
• FastAPI inference server included when --deployment fastapi

PyTorch

• Configurable nn.Sequential model with Dropout
• Full training loop with DataLoader, early stopping
• torch.save / torch.load model persistence

TensorFlow / Keras

• keras.Sequential with BatchNorm and Dropout
• EarlyStopping callback pre-configured
• model.save() / tf.keras.models.load_model() persistence

---

📊 Experiment Tracking

MLflow

mlflow.set_tracking_uri("http://localhost:5000")
with mlflow.start_run():
    mlflow.log_params(config["model"])
    mlflow.log_metrics({"accuracy": 0.95})

W&B

wandb.init(project="my-project")
wandb.log({"loss": 0.12, "accuracy": 0.95})
wandb.finish()

---

🚀 Deployment Options

FastAPI

uvicorn src.inference:app --reload
# → http://localhost:8000/docs

Docker

docker build -t my-ml-project .
docker run -p 8000:8000 my-ml-project

Kubernetes

kubectl apply -f k8s/

---

📈 Monitoring

Evidently AI

from evidently.report import Report
from evidently.metric_preset import DataDriftPreset

report = Report(metrics=[DataDriftPreset()])
report.run(current_data=current, reference_data=reference)

---

🤖 CI/CD Integration

GitHub Actions

name: Generate ML Project
on:
  workflow_dispatch:
    inputs:
      framework:
        type: choice
        options: [sklearn, pytorch, tensorflow]
      project_name:
        type: string
        default: ml-project

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v4
        with:
          python-version: '3.11'
      - run: pip install mlops-project-generator
      - run: |
          mlops-project-generator init \
            --framework ${{ github.event.inputs.framework }} \
            --project-name ${{ github.event.inputs.project_name }} \
            --tracking mlflow \
            --deployment docker
      - uses: actions/upload-artifact@v3
        with:
          name: ${{ github.event.inputs.project_name }}
          path: ${{ github.event.inputs.project_name }}/

GitLab CI

generate_ml_project:
  image: python:3.11
  script:
    - pip install mlops-project-generator
    - mlops-project-generator init \
        --framework $FRAMEWORK \
        --project-name $PROJECT_NAME \
        --tracking mlflow \
        --deployment docker
  artifacts:
    paths: [$PROJECT_NAME/]
    expire_in: 1 week

---

🏗️ Repository Structure

MLOps-Project-Generator/
├── generator/              # Python CLI source
│   ├── cli.py             # Typer CLI entry point
│   ├── renderer.py        # Jinja2 template renderer
│   ├── validators.py      # Input validation
│   ├── cloud_deployer.py  # Cloud template generation
│   └── ...
├── templates/              # Jinja2 project templates
│   ├── common/            # Shared across all frameworks
│   ├── sklearn/
│   ├── pytorch/
│   └── tensorflow/
├── web_ui/                 # Next.js web application
│   ├── app/
│   │   └── api/           # Next.js API routes (generator backend)
│   │       ├── generate/  # POST — start generation
│   │       ├── status/    # GET  — poll task status
│   │       ├── download/  # GET  — download ZIP
│   │       └── options/   # GET  — dropdown options
│   └── src/
│       └── lib/
│           ├── generator/ # Pure TS generator (no Python needed)
│           └── task-store.ts
├── tests/                  # Python CLI tests
├── pyproject.toml
└── README.md

---

🧪 Testing (CLI)

pytest tests/ -v
pytest tests/ --cov=generator --cov-report=html

🛠️ Development (CLI)

git clone https://github.com/NotHarshhaa/MLOps-Project-Generator.git
cd MLOps-Project-Generator
pip install -e ".[dev]"
black generator/ tests/
isort generator/ tests/
flake8 generator/ tests/
mypy generator/

---

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Add tests for new functionality
4. Submit a pull request

---

📄 License

MIT — see LICENSE for details.

---

🙏 Acknowledgments

• Typer · Beautiful Python CLI framework
• Next.js · React framework powering the Web UI
• Rich · Stunning terminal output
• archiver · ZIP generation for project downloads

---

📞 Support

• 🐛 GitHub Issues
• 💬 GitHub Discussions

---

🗺️ Roadmap

✅ Completed

• Python CLI with interactive + non-interactive modes
• Scikit-learn, PyTorch, TensorFlow project templates
• MLflow, W&B experiment tracking integration
• Multi-cloud deployment templates (AWS, GCP, Azure)
• Configuration presets and template customization
• Project analytics and validation
• Web UI — Next.js app with glassmorphism UI (v1.0.8)
• Backend migration — generator ported to pure TypeScript, no Python runtime needed on the server (v1.0.8)

🚀 Upcoming

• Additional frameworks (XGBoost, LightGBM, HuggingFace)
• Shareable project config links
• Template marketplace
• Team collaboration features

---

⭐ If you find this useful, please give it a star on GitHub!

Generated with ❤️ by MLOps Project Generator