diff options
| author | brkirch <brkirch@users.noreply.github.com> | 2023-01-06 16:42:47 -0500 |
|---|---|---|
| committer | brkirch <brkirch@users.noreply.github.com> | 2023-01-06 16:42:47 -0500 |
| commit | c18add68ef7d2de3617cbbaff864b0c74cfdf6c0 (patch) | |
| tree | 6388fd13dad83130907dd39859c069eb9e014416 /modules/sub_quadratic_attention.py | |
| parent | 848605fb654a55ee6947335d7df6e13366606fad (diff) | |
Added license
Diffstat (limited to 'modules/sub_quadratic_attention.py')
| -rw-r--r-- | modules/sub_quadratic_attention.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/modules/sub_quadratic_attention.py b/modules/sub_quadratic_attention.py index 95924d24..fea7aaac 100644 --- a/modules/sub_quadratic_attention.py +++ b/modules/sub_quadratic_attention.py @@ -1,7 +1,7 @@ # original source: # https://github.com/AminRezaei0x443/memory-efficient-attention/blob/1bc0d9e6ac5f82ea43a375135c4e1d3896ee1694/memory_efficient_attention/attention_torch.py # license: -# unspecified +# MIT License (see Memory Efficient Attention under the Licenses section in the web UI interface for the full license) # credit: # Amin Rezaei (original author) # Alex Birch (optimized algorithm for 3D tensors, at the expense of removing bias, masking and callbacks) |