Skip to content

Commit e7f4e31

Browse files
Yadan-WeiYadan Wei
andauthored
Build ec2 with MAX_JOBS=18 (#5216)
* build ec2 with MAX_JOBS=18 * revert toml * Revert "build ec2 with MAX_JOBS=18" This reverts commit cb92482. * remove unpin packages --------- Co-authored-by: Yadan Wei <[email protected]>
1 parent 5ca9f18 commit e7f4e31

File tree

5 files changed

+1
-33
lines changed

5 files changed

+1
-33
lines changed

pytorch/training/docker/2.7/py3/Dockerfile.ec2.cpu.core_packages.json

Lines changed: 0 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -21,14 +21,6 @@
2121
"tornado": {
2222
"version_specifier": ">=6.5.1",
2323
"skip": "True"
24-
},
25-
"s3torchconnector": {
26-
"version_specifier": "==1.4.3",
27-
"skip": "True"
28-
},
29-
"accelerate": {
30-
"version_specifier": "==1.10.1",
31-
"skip": "True"
3224
},
3325
"thinc": {
3426
"version_specifier": "==8.3.4"

pytorch/training/docker/2.7/py3/Dockerfile.sagemaker.cpu.core_packages.json

Lines changed: 0 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -18,14 +18,6 @@
1818
"version_specifier": "==0.2.4",
1919
"skip": "True"
2020
},
21-
"s3torchconnector": {
22-
"version_specifier": "==1.4.3",
23-
"skip": "True"
24-
},
25-
"accelerate": {
26-
"version_specifier": "==1.10.1",
27-
"skip": "True"
28-
},
2921
"thinc": {
3022
"version_specifier": "==8.3.4"
3123
},

pytorch/training/docker/2.7/py3/cu128/Dockerfile.ec2.gpu.core_packages.json

Lines changed: 0 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -22,14 +22,6 @@
2222
"version_specifier": "==2.3",
2323
"skip": "True"
2424
},
25-
"s3torchconnector": {
26-
"version_specifier": "==1.4.3",
27-
"skip": "True"
28-
},
29-
"accelerate": {
30-
"version_specifier": "==1.10.1",
31-
"skip": "True"
32-
},
3325
"thinc": {
3426
"version_specifier": "==8.3.4"
3527
},

pytorch/training/docker/2.7/py3/cu128/Dockerfile.gpu

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -89,7 +89,7 @@ RUN cd /tmp \
8989
ENV NVTE_FRAMEWORK=pytorch
9090
# Install flash-attn using instructions from https://github.com/Dao-AILab/flash-attention#installation-and-features
9191
# Set MAX_JOBS=4 to avoid OOM issues in installation process
92-
RUN MAX_JOBS=4 pip install --no-cache-dir flash-attn==${FLASH_ATTN_VERSION} --no-build-isolation
92+
RUN MAX_JOBS=18 pip install --no-cache-dir flash-attn==${FLASH_ATTN_VERSION} --no-build-isolation
9393
# Install TE using instructions from https://docs.nvidia.com/deeplearning/transformer-engine/user-guide/installation.html
9494
RUN pip install --no-cache-dir git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TE_VERSION} --no-build-isolation
9595

pytorch/training/docker/2.7/py3/cu128/Dockerfile.sagemaker.gpu.core_packages.json

Lines changed: 0 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -22,14 +22,6 @@
2222
"version_specifier": "==2.3",
2323
"skip": "True"
2424
},
25-
"s3torchconnector": {
26-
"version_specifier": "==1.4.3",
27-
"skip": "True"
28-
},
29-
"accelerate": {
30-
"version_specifier": "==1.10.1",
31-
"skip": "True"
32-
},
3325
"thinc": {
3426
"version_specifier": "==8.3.4"
3527
},

0 commit comments

Comments
 (0)