Service Classification based on Service Description

Forecasting Bitcoin Prices via ARIMA, XGBoost, Prophet, and LSTM models in Python

This project aims to study the Image Colorization problem and implement a Convolutional Neural Network that is able to colorize black and white images using CIELAB color space.

Specialized LSTM & AI models

Music generation using a Long Short-Term Memory (LSTM) neural network. The gennhausser project uses TensorFlow and music21 libraries to create a synthetic dataset, train an LSTM model, and generate music sequences.

Using Deep Learning to Categorize Music through Spectrogram Analysis

🚀 Unveiling Stock Market Insights with RNNs: A concise exploration of LSTM and GRU models for stock price prediction, featuring a research paper and Jupyter Notebook. 💹📈

A computer vision model for Indian Sign Language Recognition

LSTM and all other supporting modules are used to predict the next word based on the previous five words.

This repository contains a project aimed at predicting Tesla's stock prices using Long Short-Term Memory (LSTM) networks.

Machine learning pipeline for training ARIMA and LSTM models to forecast daily market prices of food products in Ethiopia. Powers the Bazarya price alert system.

The goal of this project is to accurately predict the future closing value of a given stock across a given period of time in the future.

Many-to-one LSTM neural network for binary sentiment classification of IMDB movie reviews. Built with TensorFlow/Keras as part of Deep Learning coursework. Includes data preprocessing, model training, evaluation, and visualization.

Repo for the Deep Learning Specialization offered by Coursera

A predictive analytics system for the apparel industry that forecasts daily absenteeism using LSTM/XGBoost and classifies employee risk levels to optimize workforce planning. Built with Python and Streamlit.

Create Music with Machine Learning!

Minimal Docker container for TesseractOCR

Сентиментальный анализ рынка акции

Autoencoders for vision and NLP tasks. Vision autoencoders use fully connected and convolutional architectures with layer-inverse constraints. NLP autoencoder employs LSTM-based sequence-to-sequence model for text denoising.