This repository serves as a refresher for practicing widely used Machine Learning (ML) algorithms. It includes Python implementations for various ML models, ranging from basic to advanced, using diverse datasets.
This repository is designed to reinforce concepts and skills in Machine Learning by working on hands-on examples with Python. It includes code for:
- Basic Python concepts.
- Data preprocessing.
- Supervised and unsupervised ML algorithms.
- Evaluating models and tuning hyperparameters.
Before running the code, ensure you have:
- Python 3.7+
- Libraries such as
numpy,pandas,scikit-learn,matplotlib, andseaborn.
Install dependencies using:
pip install -r requirements.txt-
Linear Regression
- Predicting house prices using the Boston Housing dataset.
-
Logistic Regression
- Classifying diabetes using the Pima Indians Diabetes dataset.
-
Decision Trees
- Predicting survival on the Titanic dataset.
-
Random Forest
- Feature importance and classification on a custom dataset.
-
K-Nearest Neighbors (KNN)
- Classification of the Iris dataset.
-
Support Vector Machines (SVM)
- Classification using a two-class synthetic dataset.
-
K-Means Clustering
- Customer segmentation on a retail dataset.
-
Principal Component Analysis (PCA)
- Dimensionality reduction on a wine quality dataset.
-
Neural Networks (Basic)
- Binary classification using a synthetic dataset.
-
Gradient Boosting (XGBoost)
- Predicting loan default on a financial dataset.
- Synthetic Datasets: Generated using
sklearn.datasetsfor SVM and Neural Networks.
- Clone this repository:
git clone https://github.com/sjat02/lambda.git
- Navigate to the directory:
cd lambda - Install required libraries:
pip install -r requirements.txt
- Explore the notebooks and run examples:
jupyter notebook
- Datasets sourced from Kaggle and
sklearn.datasets. - Tutorials and guides referenced from Scikit-learn Documentation.