ujjax/pred-rnn

PredRNN: Recurrent Neural Networks for Predictive Learning using Spatiotemporal LSTMs

40
/ 100
Emerging

Implements spatiotemporal LSTM cells that decouple spatial and temporal feature extraction through separate convolutional pathways, enabling efficient video frame prediction. Built on PyTorch with training pipelines for Moving MNIST and custom video datasets. The architecture stacks multiple ST-LSTM layers to capture both spatial transformations and temporal dynamics in sequential visual data.

127 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 9 / 25
Community 21 / 25

How are scores calculated?

Stars

127

Forks

39

Language

Python

License

MIT

Last pushed

Oct 09, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ujjax/pred-rnn"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.