Machine Learning Assignment

This repository contains the implementation of a machine learning assignment divided into two parts. The datasets used in this project are train.csv for training and test.csv for testing. The workflow includes preprocessing, exploratory data analysis, and applying a variety of machine learning models.

Part 1: Preprocessing, EDA and Models

The initial stage involves performing exploratory data analysis (EDA) and preprocessing on the training and test datasets. This step is implemented in the following file:

• PreProcessing_And_EDA.ipynb
  • Analyzes the datasets to identify patterns, missing values, and outliers.
  • Applies preprocessing steps such as handling missing data, feature engineering, and scaling.
  • Outputs the processed datasets as Processed_train.csv and Processed_test.csv, which are used in subsequent modeling steps.

Decision Tree, KNN, Random Forest, Gradient Boosting, AdaBoost, XGBoost, and Naive Bayes

In the first part, seven models are implemented:

• Decision Tree
• K-Nearest Neighbors (KNN)
• Random Forest
• Gradient Boosting
• AdaBoost
• XGBoost
• Naive Bayes

Implementation details can be found in the respective files:

• RandomForest_AdaBoost_GradientBoosting_DecisionTrees_KNearestNeighbours.ipynb
  • Implements Decision Tree, KNN, Random Forest, Gradient Boosting, and AdaBoost.
• XGBOOST_Naive_Bayes.ipynb
  • Implements XGBoost and Naive Bayes.

Part 2: Models

Logistic Regression, SVM, and Neural Networks

In the second part, three additional models are implemented:

• Logistic Regression
• Support Vector Machines (SVM)
• Neural Networks:
  • Using PyTorch
  • Using TensorFlow

Implementation details can be found in the following files:

• LR_SVM_NeuralNetworkUsingPyTorch.ipynb
  • Implements Logistic Regression, SVM, and Neural Networks using PyTorch.
• NeuralNetworkUsingTensorFlow.ipynb
  • Implements Neural Networks using TensorFlow.

License

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