Machine Learning Projects Repository

Overview

This repository contains a collection of machine learning projects covering deep learning, optimization, and classical ML algorithms. Each project demonstrates a different application of machine learning, from natural language processing to image classification and traditional classification techniques.

Projects

1. Tweet Generation with RNN (Keras)

Uses a recurrent neural network (RNN) built with Keras to generate tweets based on user input.

Trains on tweets from different users to produce contextually similar text.

2. CNN for MNIST Digit Classification

Implements a convolutional neural network (CNN) to classify digits from the MNIST dataset.

One model predicts whether an image represents the digit 7, while another classifies all ten digits (0-9).

3. LSTM for Name Prediction

Uses an LSTM-based RNN to predict names from a dataset (names.txt).

The model can be trained on different text files to generate new predictions.

4. Gradient Descent from Scratch

Implements gradient descent optimization on two functions of variables x1 and x2.

Demonstrates fundamental optimization techniques without external libraries.

5. Classification with scikit-learn

Implements various classification models using scikit-learn, including:

K-Nearest Neighbors (KNN)

Support Vector Machines (SVM) (Linear & RBF)

Decision Trees

Random Forest

Neural Networks

AdaBoost

6. KNN on the Iris Dataset

Uses the K-Nearest Neighbors (KNN) algorithm to classify flowers in the iris dataset.

Demonstrates basic supervised learning techniques.

Requirements

To run these projects, install the following dependencies:

pip install numpy pandas scikit-learn tensorflow keras matplotlib

Usage

Each project is contained in its own script or notebook.