RedML

A Python package for machine learning models built from scratch, including Linear Regression and K-Nearest Neighbors Regression. RedML is designed for learning, experimentation, and lightweight use cases.

---

Features

• Linear Regression:

  • Train and predict using gradient descent.
  • Supports single and multi-feature datasets.
  • Visualize the regression line for single-feature datasets.
  • Evaluate using the R² score.
  • Includes feature scaling for better convergence.

• K-Nearest Neighbors Regression:

  • Predict continuous target values using ( k )-nearest neighbors.
  • Supports single and multi-feature datasets.
  • Visualize regression curves for single-feature datasets.
  • Evaluate using the R² score.
  • Configurable distance metrics (e.g., Euclidean, Manhattan).

• Support Vector Regression (SVR):

  • Train using the epsilon-insensitive loss function.
  • Supports linear, polynomial, and RBF kernels.
  • Handles single and multi-feature datasets.
  • Visualize the regression fit for single-feature datasets.
  • Evaluate using the R² score.
  • Configurable regularization parameter (C) and kernel-specific parameters (e.g., gamma for RBF).

• Polynomial Regression:

  • Train and predict using gradient descent optimization.
  • Supports single-feature datasets.
  • Visualize the polynomial regression fit for single-feature datasets.
  • Configurable polynomial degree.
  • Evaluate using the R² score.
  • Automatically generates polynomial features for input data.

---

Installation

Install RedML using pip:

pip install RedML

Or clone the repository:

git clone https://github.com/Jackhammer9/RedML.git
cd RedML
pip install .

---

Usage

1. Linear Regression

Example Usage

from RedML import LinearRegressionClassifier

# Training data
X = [[1], [2], [3], [4], [5]]  # Single feature dataset
y = [1.5, 2.5, 3.5, 4.5, 5.5]

# Initialize the model
model = LinearRegressionClassifier(X, y, learningRate=0.01, maxIter=1000)

# Train the model
model.fit()

# Predict target values
X_test = [[6], [7]]
predictions = model.predict(X_test)
print("Predictions:", predictions)

# Visualize the regression line (for single-feature datasets)
model.visualize()

# Evaluate the model using R² score
r2_score = model.score(X, y)
print("R² Score:", r2_score)

---

2. K-Nearest Neighbors Regression

Example Usage

from RedML import KNNRegressionClassifier

# Training data
X = [[1], [2], [3], [4], [5]]  # Single feature dataset
y = [1.5, 2.5, 3.5, 4.5, 5.5]

# Initialize the model
knn = KNNRegressionClassifier(X, y, k=2)

# Fit the model
knn.fit()

# Predict target values
X_test = [[2.5], [3.5]]
predictions = knn.predict(X_test)
print("Predictions:", predictions)

# Visualize the regression curve (for single-feature datasets)
knn.visualize()

# Evaluate the model using R² score
r2_score = knn.score(X, y)
print("R² Score:", r2_score)

---

3. Polynomial Regression

Example Usage

from RedML import PolynomialRegressionClassifier

# Training data
X = [[1], [2], [3], [4], [5]]  # Single feature dataset
y = [2.3, 3.1, 4.8, 8.5, 10.1]

# Initialize the model
model = PolynomialRegressionClassifier(X, y, degree=3, learningRate=0.0001, maxIter=3000)

# Train the model
model.fit()

# Predict target values
X_test = [[6]]
predictions = model.predict(X_test)
print("Predictions:", predictions)

# Visualize the polynomial regression curve
model.visualize()

# Evaluate the model using R² score
r2_score = model.score(X, y)
print("R² Score:", r2_score)

---

4. Support Vector Regression

Example Usage

from RedML import EpsilonSVRClassifier

    # Training data
    X = np.array([[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]])
    y = np.array([2.3, 3.1, 4.8, 8.5, 10.1, 11.8, 13.2, 14.4, 16, 17.6])

    # Initialize and train the SVR model
    svr = EpsilonSVRClassifier(X, y, epsilon=0.1, C=10, kernel="rbf", gamma=0.5)
    svr.fit()

    # Predict and evaluate
    X_test = np.array([[6]])
    predictions = svr.predict(X_test)
    print("Predictions on X_test:", predictions)

    # Visualize the results
    svr.visualize()

    # R^2 score
    r2_score = svr.score(X, y)
    print("R^2 Score:", r2_score)

---

Contributing

Contributions are welcome! If you'd like to improve RedML, please fork the repository and submit a pull request.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.