vtu-ml-lab 1.1.0

VTU Machine Learning Lab programs — library, viewer, and exam assistant

vtu-ml-lab

vtu-ml-lab is a comprehensive Python library, interactive viewer, and command-line examination assistant tailored for VTU Machine Learning Laboratory (21CS63 / 18CSL76 / similar schemes). It is designed to help students and educators view, save, execute, and prepare for exams with inline viva questions and memory tricks.

---

Features

• All 10 Lab Programs: Ready-to-use, clean, and fully-commented implementations of standard VTU Machine Learning laboratory algorithms.
• Dual Viewer Support: Seamless rendering of code with syntax highlighting in Jupyter Notebooks (IPython.display.Code) and standard outputs in standard terminals.
• Code Saver: Export any lab script instantly to a file, automatically building any missing parent directories.
• Isolated Execution: Execute lab code locally in isolated namespaces, with options to prompt for confirmations or run non-interactively (Google Colab / script environments).
• Exam Assistant / Viva Prep: Embedded metadata including 6–8 inline Viva Q&As per lab, step-by-step logic, and specific memory mnemonics for memorization.
• Interactive CLI: Rich, argument-parsed command-line utility with shortcuts.

---

Project Structure

File / Folder	Description
vtu_ml_lab/	Core package folder
├── __init__.py	Package entry point, exposing standard APIs
├── programs.py	Central database containing code, titles, and metadata for all 10 programs
├── viewer.py	Code retrieval, listing, and Jupyter-sensitive code display helpers
├── saver.py	Disk storage helper with automatic directory creation
├── runner.py	Clean python execution engine using isolated dictionary scope namespaces
├── metadata.py	Query helpers for Viva lists, logic guides, memory tricks, and expected outputs
└── cli.py	Command line routing, supporting argument parsing and digit shortcuts
tests/	Unit testing suites
├── test_vtu_ml_lab.py	Complete pytest suites verifying API behavior and metadata sanity
pyproject.toml	Standard modern build setup, dependencies, and script routing
setup.py	Minimal legacy compatibility wrapper
requirements.txt	Clean development dependency lists
LICENSE	MIT License

---

Installation

Install directly from PyPI:

pip install vtu-ml-lab

For development or local setup:

git clone https://github.com/prathamb/vtu-ml-lab.git
cd vtu-ml-lab
pip install -r requirements.txt
pip install -e .

Dependencies

• scikit-learn >= 1.3
• pandas >= 2.0
• matplotlib >= 3.7
• seaborn >= 0.12
• numpy >= 1.24
• statsmodels >= 0.14

---

Python API Usage

Here is how you can use vtu-ml-lab programmatically within your Python scripts or Jupyter Notebooks:

import vtu_ml_lab as ml

# 1. List all available lab programs
ml.list_programs()

# 2. Get program code as a string
code_str = ml.get_program(3)
print(code_str[:200])

# 3. View program inside terminal or Jupyter
ml.show_program(3)

# 4. Save program to disk (defaults to lab<N>.py if path not specified)
ml.save_program(3, "my_labs/lab3_pca.py")

# 5. Run the lab program (confirm=True prompts the user first)
ml.run_program(3, confirm=False)

# 6. Retrieve Exam Study Material
print("Viva Q&As:", ml.get_viva(3))
print("Memory Mnemonic:", ml.get_memory_trick(3))
print("Key Lines of Code:", ml.get_important_lines(3))
print("Program Logic:", ml.get_logic(3))
print("Expected Output:", ml.get_output(3))

---

CLI Usage

The package exposes a vtu-ml-lab CLI script when installed.

List available programs:

vtu-ml-lab list

Display code for a program:

vtu-ml-lab show 3

Or use the bare integer shortcut:

vtu-ml-lab 3

Save code to file:

# Saves as lab3.py in current directory
vtu-ml-lab save 3

# Saves to custom path, automatically creating parent directories
vtu-ml-lab save 3 my_workspace/lab3_iris_pca.py

Execute a program:

# Prompts for confirmation before running
vtu-ml-lab run 3

# Runs directly without interactive confirmation (ideal for scripts/Colab)
vtu-ml-lab run 3 --yes

Get Viva Q&As:

vtu-ml-lab viva 3

Get Memory Trick:

vtu-ml-lab trick 3

Get Step-by-step Logic:

vtu-ml-lab logic 3

Get Key Lines of Code:

vtu-ml-lab lines 3

Get Expected Output Details:

vtu-ml-lab output 3

---

PyPI Publishing Steps

Follow these steps to build and upload the package to PyPI:

1. Verify Local Setup: Ensure you have build and twine installed:

   pip install --upgrade build twine
   

2. Run Tests: Ensure all package tests are passing:

   pytest tests/ -v
   

3. Build the Distribution Packages: Build the source distribution and wheel:

   python -m build
   

   This will create a dist/ directory containing .tar.gz and .whl files.

4. Verify Build Contents: Verify that files are packaged correctly:

   twine check dist/*
   

5. Upload to TestPyPI (Optional but Recommended): Upload to the TestPyPI registry to verify packaging:

   python -m twine upload --repository testpypi dist/*
   

   Provide your TestPyPI API token when prompted.

6. Upload to PyPI: Upload the final package to the live PyPI registry:

   python -m twine upload dist/*
   

   Provide your PyPI API token when prompted.

---

License

This project is licensed under the MIT License - see the LICENSE file for details. Author: Pratham Balehosur