minGRU
Project description
minGRU
Implementation of the proposed minGRU in Pytorch, only the log-space numerically stable version.
Install
$ pip install minGRU-pytorch
Usage
import torch
from minGRU_pytorch import minGRU
min_gru = minGRU(512)
x = torch.randn(2, 1024, 512)
out = min_gru(x)
assert x.shape == out.shape
Test
enwik8
$ python train.py
Citations
@inproceedings{Feng2024WereRA,
title = {Were RNNs All We Needed?},
author = {Leo Feng and Frederick Tung and Mohamed Osama Ahmed and Yoshua Bengio and Hossein Hajimirsadegh},
year = {2024},
url = {https://api.semanticscholar.org/CorpusID:273025630}
}
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
mingru_pytorch-0.0.3.tar.gz
(36.6 MB
view details)
Built Distribution
File details
Details for the file mingru_pytorch-0.0.3.tar.gz
.
File metadata
- Download URL: mingru_pytorch-0.0.3.tar.gz
- Upload date:
- Size: 36.6 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.9.20
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | dbd969d9256ca83efae10af5d4ff5235039e612964ddece3eee0710dbcdc7679 |
|
MD5 | 5530c03a213aaee70c7320fc8821d136 |
|
BLAKE2b-256 | 800c3ba6cb78bbc0c4d5cc8a1fa68a955e8757f56cef00b2fc5b337fd0a9aaec |
File details
Details for the file mingru_pytorch-0.0.3-py3-none-any.whl
.
File metadata
- Download URL: mingru_pytorch-0.0.3-py3-none-any.whl
- Upload date:
- Size: 4.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.9.20
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 598ce79a7785f1df0c8f16515f684cd0bff23020f3fee8034a6f0542df3e5589 |
|
MD5 | 5bb8da67ea115758b2192265b1c9bb8f |
|
BLAKE2b-256 | 42f3c7ded794619fc1d1f303df5721c72a54087a7c56fa930c1cb0643c4d87b1 |