A combined LSTM and LightGBM framework for improving deterministic and probabilistic wind energy forecasting

Runner-up team (2nd place) in AI4VN2022: Air Quality Forcasting Challenge

[TOIS] "Privacy-Preserving Individual-Level COVID-19 Infection Prediction via Federated Graph Learning"

StockLLM: A Stock Analyzer with Comprehensive LLM Insights

weatheril is an unofficial [IMS](https://ims.gov.il) (Israel Meteorological Service) python API wrapper.

Scripts to build and run WRF and WPS in docker

OmniEcon Nexus is an open-source, high-performance simulation engine for global micro/macro-economic analysis. Using deep learning, agent-based modeling, and optimization, it supports 5M agents for forecasting, risk analysis, policy simulation, and portfolio management. Built for governments, researchers, and developers.

An implementation of AE LSTM based. We test our architecture on several tasks as reconstructing synthetic time series, s&p 500 stocks, and forecasting s&p 500 stocks based on the decoded information (also known as latent space) features we extract from the AE

This project involves developing and testing a trading model designed to predict stock prices and evaluate trading strategies. The core of the project includes building and training a LSTM based model for time series forecasting in addition to a RL model, evaluating its performance, and visualizing the results.

forecasting time series Singapore PSI (pm2.5) 2016-2019

This Model is Base On Halt & Winter Algorithm.This Model is Forecast About Seasonal Data.

The Alpha Alternator is a novel generative model designed for time-dependent data, dynamically adapting to varying noise levels in sequences.

A light-code version of Time-LLM based on GPT2

CryptoForecasting flask project aimed at predicting cryptocurrency prices for Bitcoin (BTC) and Ethereum (ETH) using machine learning and deep learning.

APMLV ( i.e. Automated Prediction and Management of Logical Volumes ) is a project that leverages deep learning and automation to optimize the management of logical volumes resources

The random walk application receives a number of steps from the user and simulates a walk in a random direction with equal step sizes (one unit). Following each run, a histogram plots the distances from the origin for each run, and the expected value can be evaluated as the average distance approach a certain value.

A plug and play framework for Temporal Fusion Transformer. Predict your future!

RNN-LSTM (Recurrent Neural Networks based Long Short-Term Memory). Designed to train a model for processing DNA/RNA sequences.

Algorithmic forecasting of energy consumption

NFV System Supporting Autoscaling of Bare-metal Resource