OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning

Official implementation for NIPS'17 paper: PredRNN: Recurrent Neural Networks for Predictive Learning Using Spatiotemporal LSTMs.

PredRNN: Recurrent Neural Networks for Predictive Learning using Spatiotemporal LSTMs

MultiGraphGAN for predicting multiple target graphs from a source graph using geometric deep learning.

Graph SuperResolution Network using geometric deep learning.

Analysis code for the OpenScope Credit Assignment project.

Predicting multigraph brain population from a single graph

Code release for "PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning" (ICML 2018)

A Python toolbox for predicting brain network (graph) evolution over time from a single observation. The codes of the 20 competing Kaggle teams along with the competition datasets are made available.

ABMT (Adversarial Brain Multiplex Translator) for brain graph translation using geometric generative adversarial network (gGAN).

Brain Graph Super-Resolution: how to generate high-resolution graphs from low-resolution graphs? (Python3 version)

Federated time-dependent graph evolution prediction with missing timepoints.

Our group project for Govhack2023

A few-shot learning approach to forecasting the evolution of the brain connectome.

PredRNN implementation using Tensorflow.

Generative Predictive Networks — an experimental attempt to stabilize GANs' training.

One Algorithm, Two Models, and a Prediction

Simple prolog application that uses predicate logic to diagnose diseases.