A beginner-friendly machine learning library that automates preprocessing, training, and evaluation.

This is Python version of this library visit more.

• C++ Version

Origination - rslearn-lib

• rslearn-ML
• rslearn-lib

• ⚡ Minimal setup — no complex configuration
• 🤖 Automatic pipeline (scaling, splitting, evaluation)
• 📊 Built-in metrics for regression & classification
• 🧠 Designed for beginners learning ML concepts
• 🧩 Clean and simple API inspired by sklearn

• Version : 1.0.8 - 1.0.3
• Release Date: 2026-06-26

• KNN Regression & Classification Both
• Evaluation & auto scaling support

More Info: CHANGELOG

Downloads - Module

• Linear Regression (Single & Multi-feature)
• Logistic Regression (Binary & Multi-class)
• Ridge Regression (L2 Regularization)
• Lasso Regression (L1 Regularization)
• Elastic Net (L1 + L2)

• Mean Squared Error (MSE)
• Mean Absolute Error (MAE)
• Root Mean Squared Error (RMSE)
• R² Score
• Accuracy (for classification)

✔ Supports single-output and multi-output tasks

• StandardScaler
• MinMaxScaler

• Train-Test Split

  • Supports stratify for balanced sampling

All models in rslearn are implemented using Gradient Descent.

⚠️ Important:

• Feature scaling is highly recommended for stable and faster convergence.

• Use:

  • StandardScaler (recommended)
  • or MinMaxScaler

models include Inbuilt StandardScaler Feature in fit() Method:

scale=True  # default

• Automatically applies feature scaling internally
• Helps prevent numerical instability

rslearn/
│
├── linear_model/
│   ├── _linear_regression.py
│   ├── _logistic_regression.py
│   ├── _ridge.py
│   ├── _lasso.py
│   ├── _elastic_net.py
│
├── preprocessing/
│   ├── _scaler.py
│
├── metrics/
│   ├── _regression.py
│
├── model_selection/
│   ├── _split.py
│
└── README.md

📌 Each module contains its own detailed README with usage examples and explanations.

git clone https://github.com/rslearn-lib/rslearn-ML-py.git
cd rslearn-ML-py/

pip install rslearn-py

Downloads Older Library

pip install -r requirements.txt

from rslearn.linear_model import LinearRegression
from rslearn.preprocessing import StandardScaler
import numpy as np

X = np.array([10, 20, 30])
y = np.array([5, 10, 15])

scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

model = LinearRegression()
model.fit(X_scaled, y)

print(model.predict([40]))

• Each folder includes its own README.md

• Covers:

  • Usage
  • Parameters
  • Examples
  • Internal working

• Understand ML algorithms from scratch
• Build a sklearn-like API
• Create reusable and modular ML components
• Learn real-world ML system design
• Check Self Ability

ItzRustam

rslearn-lib

This project is licensed under the GNU GPL v3 License.