MedGAN is a repository of GAN implementations for generating high-quality, diverse, and stable medical images, focusing on brain tumor MRI scans.
The study uses a GRU neural network with multiple layers and a dense output layer, trained with the Adam optimizer and MSE loss for time-series prediction.
The project compares Logistic Regression, KNN, and SVC for phishing website detection, with SVC optimized using GridSearch and evaluated on accuracy, F1 score, recall, and precision.
This project implements a Speech-to-Text system using Word2Vec for enhanced accuracy, with audio processed into text and deployed on a web platform using Flask, HTML, and CSS.
The Voice Emotion Classification project employs an LSTM model with MFCC features to classify emotions like fear and neutrality.
This project uses YOLOv3 for real-time object detection, classifying and localizing objects in uploaded images. Deployed via Streamlit, it offers an interactive web interface for instant and practical testing.
Arduino robot with circular attack and smart movement.
This project implements a recommendation system using Singular Value Decomposition (SVD) with the surprise library for movie/book/user preference predictions and collaborative filtering.
This project uses a CNN with data augmentation to classify satellite images into four categories with high accuracy. The model was deployed via a Telegram bot, enabling real-time image classification and showcasing the power of deep learning in remote sensing.
This project implements an anomaly detection system for synthetic network traffic using an Isolation Forest model and deep learning neural network to identify potential security threats or unusual activity.
This project predicts customer churn using machine learning models and an Artificial Neural Network (ANN) to help businesses retain valuable customers.
This project implements a deep learning-based brain tumor detection system using VGG16, MobileNetV3, and InceptionV3 to classify brain MRI images as Tumor or Healthy.
This project implements sentiment analysis on IMDB movie reviews using LSTM and GRU architectures to classify reviews as positive or negative.
This project builds an emotion recognition system using CNN and OpenCV to classify facial expressions into emotions, enhancing human-computer interaction with real-time video stream detection.
This repository contains supervised and unsupervised machine learning models, showcasing algorithms for classification, regression, clustering, and dimensionality reduction, with practical implementations for real-world datasets.
This repository guides predicting Netflix's stock closing prices using LSTM, GRU, and Bidirectional LSTM to explore their efficacy in time series forecasting.
This repository demonstrates image classification using transfer learning with VGG16, InceptionV3, and MobileNet, leveraging pre-trained models for efficient classification with fewer data and resources.