PyTorch implementations of the paper, 'Convolutional Tensor-Train LSTM for Spatio-Temporal Learning', NeurIPS 2020. [project page]
- code/ (original): The original implementation of the paper.
- code_opt/ (optimized): The optimized implementation to accelerate training.
Copyright (c) 2020 NVIDIA Corporation. All rights reserved. This work is licensed under a NVIDIA Open Source Non-commercial license.
- Moving-MNIST-2
- KTH action
Higher PSNR/SSIM and lower MSE/LPIPS values indicate better predictive results. # of FLOPs denotes the multiplications for one-step prediction per sample, and Time(m) represents the clock time (in minutes) required by training the model for one epoch (10,000 samples)
This code was written by Wonmin Byeon ([email protected]) and Jiahao Su ([email protected]).