ml-lab-toolkit 1.0.0

A professional Python library implementing core machine learning algorithms from ML lab experiments.

ML Lab Toolkit

A production-ready Python library implementing core machine learning algorithms from ML lab experiments. Install worldwide with:

pip install ml-lab-toolkit

Installation

pip install ml-lab-toolkit

Development install:

git clone https://github.com/your-username/ml-lab-toolkit.git
cd ml-lab-toolkit
pip install -e ".[dev]"

Quick Start

from mllab import PCA, KNN, FindS, LWR
from mllab import LinearRegression, PolynomialRegression
from mllab import DecisionTree, NaiveBayes, KMeans
import numpy as np

X = np.random.randn(50, 4)
pca = PCA(n_components=2)
X_reduced = pca.fit_transform(X)

knn = KNN(n_neighbors=3)
knn.fit(X[:30], np.random.randint(0, 2, 30))
predictions = knn.predict(X[30:])

Algorithms

Module	Class	Description
mllab.pca	PCA	Principal Component Analysis
mllab.find_s	FindS	Find-S concept learning
mllab.knn	KNN	k-Nearest Neighbors classifier
mllab.lwr	LWR	Locally Weighted Regression
mllab.linear_regression	LinearRegression	Ordinary least squares regression
mllab.polynomial_regression	PolynomialRegression	Polynomial feature regression
mllab.decision_tree	DecisionTree	Entropy-based decision tree
mllab.naive_bayes	NaiveBayes	Gaussian Naive Bayes
mllab.kmeans	KMeans	K-Means clustering

Usage Examples

PCA (Experiment 3)

from mllab import PCA
import numpy as np

X = np.random.randn(150, 4)
model = PCA(n_components=2, standardize=True)
X_2d = model.fit_transform(X)
print(model.explained_variance_ratio_)

Find-S (Experiment 4)

from mllab import FindS

model = FindS(positive_label="Yes")
model.fit(X_attributes, y_labels)
print(model.hypothesis_)

KNN (Experiment 5)

from mllab import KNN

model = KNN(n_neighbors=5)
model.fit(X_train, y_train)
accuracy = model.score(X_test, y_test)

LWR (Experiment 6)

from mllab import LWR

model = LWR(tau=1.0)
model.fit(x_train, y_train)
y_pred = model.predict(x_test)

Linear & Polynomial Regression (Experiment 7)

from mllab import LinearRegression, PolynomialRegression

linear = LinearRegression().fit(X, y)
poly = PolynomialRegression(degree=3).fit(X, y)

Decision Tree (Experiment 8)

from mllab import DecisionTree

tree = DecisionTree(max_depth=7)
tree.fit(X_train, y_train)
predictions = tree.predict(X_test)

Naive Bayes (Experiment 9)

from mllab import NaiveBayes

nb = NaiveBayes().fit(X_train, y_train)
accuracy = nb.score(X_test, y_test)

K-Means (Experiment 10)

from mllab import KMeans

kmeans = KMeans(n_clusters=2, random_state=42)
labels = kmeans.fit_predict(X)

API Documentation

All estimators follow a scikit-learn-like API:

• fit(X, y) — train the model
• predict(X) — generate predictions
• score(X, y) — evaluate performance (where applicable)
• fit_transform(X) — fit and transform (PCA)
• fit_predict(X) — fit and return labels (KMeans)

Shared utilities live in mllab.utils:

• StandardScaler
• as_2d_array, as_1d_array
• euclidean_distances

Project Structure

ml-lab-toolkit/
├── src/mllab/
│   ├── pca.py
│   ├── find_s.py
│   ├── knn.py
│   ├── lwr.py
│   ├── linear_regression.py
│   ├── polynomial_regression.py
│   ├── decision_tree.py
│   ├── naive_bayes.py
│   ├── kmeans.py
│   └── utils/
├── examples/
├── tests/
├── pyproject.toml
├── requirements.txt
└── README.md

Running Tests

pytest -v

Building for PyPI

pip install build twine
python -m build
twine check dist/*

Publishing to PyPI

1. Create accounts on PyPI and TestPyPI.
2. Generate an API token.
3. Build the package:

python -m build

4. Upload to TestPyPI first:

twine upload --repository testpypi dist/*
pip install -i https://test.pypi.org/simple/ ml-lab-toolkit

5. Upload to PyPI:

twine upload dist/*

6. Verify installation:

pip install ml-lab-toolkit
python -c "from mllab import PCA, KNN; print('OK')"

License

MIT License. See LICENSE.