Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
flash-attention
Commits
e43a4cea
"src/include/blockwise_gemm.hpp" did not exist on "c866773642cf7cf8386303b6167844d93e608547"
Commit
e43a4cea
authored
Jan 21, 2024
by
Tri Dao
Browse files
[CI] Fix CUDA 12.2.2 compilation
parent
f9d73761
Changes
3
Show whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
5 additions
and
5 deletions
+5
-5
.github/workflows/publish.yml
.github/workflows/publish.yml
+3
-3
flash_attn/__init__.py
flash_attn/__init__.py
+1
-1
training/Dockerfile
training/Dockerfile
+1
-1
No files found.
.github/workflows/publish.yml
View file @
e43a4cea
...
...
@@ -64,11 +64,11 @@ jobs:
python-version
:
'
3.7'
# Pytorch <= 2.0 only supports CUDA <= 11.8
-
torch-version
:
'
1.12.1'
cuda-version
:
'
12.2.
0
'
cuda-version
:
'
12.2.
2
'
-
torch-version
:
'
1.13.1'
cuda-version
:
'
12.2.
0
'
cuda-version
:
'
12.2.
2
'
-
torch-version
:
'
2.0.1'
cuda-version
:
'
12.2.
0
'
cuda-version
:
'
12.2.
2
'
steps
:
-
name
:
Checkout
...
...
flash_attn/__init__.py
View file @
e43a4cea
__version__
=
"2.4.3"
__version__
=
"2.4.3
.post1
"
from
flash_attn.flash_attn_interface
import
(
flash_attn_func
,
...
...
training/Dockerfile
View file @
e43a4cea
...
...
@@ -85,7 +85,7 @@ RUN pip install transformers==4.25.1 datasets==2.8.0 pytorch-lightning==1.8.6 tr
RUN
pip
install
git+https://github.com/mlcommons/logging.git@2.1.0
# Install FlashAttention
RUN
pip
install
flash-attn
==
2.4.3
RUN
pip
install
flash-attn
==
2.4.3
.post1
# Install CUDA extensions for fused dense
RUN
pip
install
git+https://github.com/HazyResearch/flash-attention@v2.4.2#subdirectory
=
csrc/fused_dense_lib
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment