This project forecasts MSFT stock prices by comparing four advanced deep learning models: TFT, TCN, DeepAR, and N-BEATS. It uses a robust pipeline with technical indicators as features. The TCN model achieved the highest accuracy, demonstrating a comprehensive approach to time-series model selection.

TCN protocol based on Private Set Intersection Cardinality.

This is the page for our semester project for CS581 - Embedded Machine Learning. CAN DO - Controller Area Network (CAN) - Discrepancy/Divergence/Deviation Observer(TBD)

Physics-Informed Graph Attention Network with Temporal Convolutional Encoder for multi-node distribution grid state forecasting under solar PV uncertainty. First benchmark for V/P/Q forecasting on IEEE 33-bus.

Sequences classification using RNN, CNN, TCN and linear models.

Full-stack AI crypto trading platform: PyTorch deep learning model (TCN) generates real-time BTC trading signals, automated execution engine with risk management, Streamlit dashboard with live analytics, FastAPI backend, SQLite ORM. End-to-end system design from ML model to production deployment in algo trading.

This repository is part of my thesis about surgical phase recognition in laparoscopic cholecystectomy. The dissertation was approved by the Department of Applied Informatics of the University of Macedonia for the attainment of Bachelor’s degree in computer science.

The static library of Train Real Time Data Protocol (TRDP)

Time series forecasting examples

fast, accurate beat estimation

Se estudia la relación entre el tipo de cambio nominal y la difusión de noticias macroeconómicas en Argentina mediante la estimación de un modelo VARX-GARCH.

Production-grade Deep Learning pipeline using Temporal Convolutional Networks (TCN) to forecast ad lead generation. Features hybrid JSON/CSV ingestion, Optuna Bayesian optimization, and probabilistic forecasting.

Applying a TCN model for a time series task

BitePulse AI - Early detection and habit coaching from bite patterns.

SelfGNNplus achieves up to 5.97% improvement in Hit Rate and 6.41% in NDCG compared to the best baseline models. Ablation studies highlight the critical role of interval-level dependencies, revealing their substantial impact on recommendation accuracy. The study also examines the effects of key hyperparameters on model performance and computation.

Uses a pose estimation model that extracts key points from each frame. This data is fed into a custom TCN before outputting predictions for three distinct infraction cards as defined by the IPF.

The platform provides comprehensive insights for making informed investment decisions through various data sources and machine learning models.

A Kotlin implementation of the TCN protocol.