labexp 1.0.0

Lab Experiments: ML experiment management for college labs and restricted servers

labexp: Lab Experiments

Fast, lightweight ML experiment management for college labs and restricted servers.

Perfect for educational institutions, lab environments, and networks with restricted internet access.

✨ Features

• 🧪 11 Pre-built ML Experiments - Ready-to-run templates (Linear Regression, SVM, Neural Networks, K-Means, etc.)
• 📊 Experiment Tracking - Store and compare ML experiments
• 🔧 Lightweight - Minimal dependencies, works offline
• 🏫 Lab-Ready - Works on restricted servers and college networks
• 📦 Zero-Config Installation - Single command setup
• 🎓 Educational - Perfect for ML courses and workshops

🚀 Quick Start

Installation

Standard (with internet):

pip install labexp

Offline/Restricted Server (no internet):

# 1. Download wheel on any machine
pip download labexp --no-deps --python-version 38 --only-binary=:all:

# 2. Transfer .whl file to lab machine

# 3. Install offline
pip install labexp-1.0.0-py3-none-any.whl

From Source (most flexible):

git clone https://github.com/yourusername/labexp.git
cd labexp
pip install -e .

Usage

import labexp as exp

# Show experiment code
exp.exp(1)              # Linear Regression
exp.exp(5)              # SVM
exp.exp(8)              # Neural Network

# List all 11 experiments
exp.list_experiments()

# Get experiment code as string
code = exp.get_experiment_code(3)

# Get experiment info
info = exp.get_experiment_info(2)

📚 Available Experiments

1. Linear Regression - Basic regression with scikit-learn
2. Logistic Regression - Binary classification
3. Decision Trees - Tree-based classification
4. Random Forest - Ensemble learning
5. Support Vector Machine - SVM classification
6. K-Means Clustering - Unsupervised learning
7. Gradient Boosting - XGBoost implementation
8. Neural Networks - Multi-layer perceptron
9. Naive Bayes - Probabilistic classifier
10. PCA - Dimensionality reduction
11. Cross-Validation - Model evaluation

💾 Offline Features

1. No Internet Required After Installation

• All 11 experiments are bundled
• All dependencies downloaded at install time
• Run experiments completely offline

2. Portable Installation

• Works on USB drives
• Can copy installed package to other machines
• Minimal size (~50 MB total)

3. Network Restrictions

• Works behind proxies
• Works on corporate/university firewalls
• No external API calls required
• Pure Python, no binary compilation needed

4. College Lab Support

• Works on Linux, Windows, macOS
• Compatible with standard Python 3.8+
• No admin privileges required
• Shared lab environment friendly

🔧 Installation Guide for Restricted Servers

For Lab Administrators

1. Download the package on any machine with internet:

   pip download labexp --no-deps
   pip download pandas
   

2. Create offline environment:

   mkdir /opt/labexp-repo
   cp *.whl /opt/labexp-repo/
   

3. Install for all users:

   pip install --no-index --find-links=/opt/labexp-repo labexp
   

4. Test installation:

   python -c "import labexp; labexp.exp(1)"
   

For Individual Users

1. Download offline installer:

   • Visit: https://pypi.org/project/labexp/
   • Download the .whl file

2. Transfer to lab machine (via USB/external drive)

3. Install:

   pip install labexp-1.0.0-py3-none-any.whl
   

4. Verify:

   python
   >>> import labexp
   >>> labexp.exp(1)
   

📖 API Reference

exp(i)

Display the code for experiment i (1-11)

labexp.exp(5)  # Shows SVM experiment code

list_experiments()

List all available experiments

labexp.list_experiments()

get_experiment_code(i)

Get experiment code as a string

code = labexp.get_experiment_code(3)

get_experiment_info(i)

Get complete experiment information

info = labexp.get_experiment_info(2)

🎓 Educational Use Cases

• ML Courses - Use as teaching resource for 11 core algorithms
• Workshops - Run experiments without internet dependency
• Lab Sessions - Pre-load on lab machines for students
• Offline Learning - Perfect for limited-connectivity environments

⚙️ System Requirements

• Python: 3.8 or higher
• OS: Linux, macOS, Windows
• Dependencies: pandas (minimal overhead)
• Optional: scikit-learn, numpy for full ML features
• Storage: ~50 MB

🛠️ Development

Install development version:

pip install -e ".[dev]"

Run tests:

pytest tests/

Build distribution:

python -m build

📋 Comparison

Feature	labexp	Standard ML Libraries
Offline Installation	✅	❌
Pre-built Experiments	✅	❌
Zero Config	✅	❌
Lab-ready	✅	⚠️
Minimal Dependencies	✅	❌

📝 License

MIT License - See LICENSE file for details

🤝 Contributing

Contributions welcome! Please submit pull requests or issues on GitHub.

📞 Support

• Documentation: See /docs folder
• Issues: GitHub Issues
• Email: [e-mail]

---

Made for education. Works everywhere.