Back to search
Top Contributors
116
commits
2
commits
Implementation of Memorizing Transformers (ICLR 2022), attention net augmented with indexing and retrieval of memories using approximate nearest neighbors, in Pytorch
Stars
641
Forks
47
Watchers
641
Open Issues
10
Overall repository health assessment
No package.json found
This might not be a Node.js project
116
commits
2
commits
address https://github.com/lucidrains/memorizing-transformers-pytorch/issues/16
272e39bView on GitHubprepare to use knn attention in another repository, for the ultimate long context attention in the world
c44f755View on GitHubUpdate KNNMemory.add to correctly index and update KNN list, after adding new memories in parallel
43cb775View on GitHubaddress https://github.com/lucidrains/memorizing-transformers-pytorch/issues/10
7f61766View on GitHubUpdate train.py to correct implementation of val loss calculation
ed58a77View on GitHubgive knn attention layer one more way to tune out local if need be
aaf9a0aView on GitHuballow the network to pay more attention to memory later into training, if need be
fabbe14View on GitHub