Skip to content

ci/gpu: setting oldest dependencies #20939

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 11 commits into from
Jun 25, 2025
Merged

ci/gpu: setting oldest dependencies #20939

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
3292c9a
ci/gpu: setting oldest dependencies
Borda Jun 25, 2025
0a2717f
pip install "cython<3.0"
Borda Jun 25, 2025
c362ab6
deepspeed >=0.9.0
Borda Jun 25, 2025
da61165
Revert "deepspeed >=0.9.0"
Borda Jun 25, 2025
fb3640c
$CUDA_HOME_PATH
Borda Jun 25, 2025
1ea404f
Revert "$CUDA_HOME_PATH"
Borda Jun 25, 2025
5dd5377
deepspeed ==0.9.3
Borda Jun 25, 2025
9d3f00a
typing-extensions >=4.5.0
Borda Jun 25, 2025
6cf54b4
PyYAML >5.4
Borda Jun 25, 2025
0aec96a
torchmetrics >0.7.0
Borda Jun 25, 2025
eb8cca3
lightning-utilities >=0.10.0
Borda Jun 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .azure/gpu-tests-fabric.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,6 +100,12 @@ jobs:
pip list
displayName: "Image info & NVIDIA"

- bash: |
python .actions/assistant.py replace_oldest_ver
pip install "cython<3.0" wheel # for compatibility
condition: contains(variables['Agent.JobName'], 'oldest')
displayName: "setting oldest dependencies"

- bash: |
PYTORCH_VERSION=$(python -c "import torch; print(torch.__version__.split('+')[0])")
pip install -q wget packaging
Expand Down
6 changes: 6 additions & 0 deletions .azure/gpu-tests-pytorch.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,6 +104,12 @@ jobs:
pip list
displayName: "Image info & NVIDIA"

- bash: |
python .actions/assistant.py replace_oldest_ver
pip install "cython<3.0" wheel # for compatibility
condition: contains(variables['Agent.JobName'], 'oldest')
displayName: "setting oldest dependencies"

- bash: |
PYTORCH_VERSION=$(python -c "import torch; print(torch.__version__.split('+')[0])")
pip install -q wget packaging
Expand Down
2 changes: 1 addition & 1 deletion requirements/fabric/base.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,5 +4,5 @@
torch >=2.1.0, <2.8.0
fsspec[http] >=2022.5.0, <2025.6.0
packaging >=20.0, <=25.0
typing-extensions >=4.4.0, <4.15.0
typing-extensions >=4.5.0, <4.15.0
lightning-utilities >=0.10.0, <0.15.0
2 changes: 1 addition & 1 deletion requirements/fabric/strategies.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,6 @@

# note: is a bug around 0.10 with `MPS_Accelerator must implement all abstract methods`
# shall be resolved by https://github.com/microsoft/DeepSpeed/issues/4372
deepspeed >=0.8.2, <=0.9.3; platform_system != "Windows" and platform_system != "Darwin" # strict
deepspeed >=0.9.3, <=0.9.3; platform_system != "Windows" and platform_system != "Darwin" # strict
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Dropping the 0.8 since it would need to be compiled from source
Also, noted that we are quite far behind the latest 0.17 🤔

Why Upgrade? Upgrading to v0.17 delivers significant performance, stability, and integration benefits—vital for training larger models with improved efficiency and reliability.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • ZeRO Optimizations: • v0.9: Early experiments in partitioning model states for memory savings. • v0.17: Advanced refinements (ZeRO-Offload and improved stage 3) enable training massively scaled models.

  • Performance Enhancements: • Upgraded distributed communication and fused operations. • Better mixed precision (fp16/bf16) support for faster training and efficient hardware usage.

  • Stability & API Maturation: • Streamlined configuration, enhanced documentation, and robust testing. • Fewer bugs and smoother integration with frameworks like HuggingFace Transformers.

  • Inference Improvements: • Expanded inference API with support for quantization. • Optimized runtime strategies for production deployment.

  • Ecosystem Integration: • Broader compatibility with modern AI tools and libraries. • Simplifies building and deploying complex deep learning workflows.

# skip bitsandbytes==0.46, due to ValueError: too many values to unpack (expected 2)
bitsandbytes >=0.45.2,!=0.46,<0.47.0; platform_system != "Darwin"
2 changes: 1 addition & 1 deletion requirements/pytorch/base.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,5 +7,5 @@ PyYAML >=5.4, <6.1.0
fsspec[http] >=2022.5.0, <2025.6.0
torchmetrics >=0.7.0, <1.8.0
packaging >=20.0, <=25.0
typing-extensions >=4.4.0, <4.15.0
typing-extensions >=4.5.0, <4.15.0
lightning-utilities >=0.10.0, <0.15.0
2 changes: 1 addition & 1 deletion requirements/pytorch/strategies.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,4 +3,4 @@

# note: is a bug around 0.10 with `MPS_Accelerator must implement all abstract methods`
# shall be resolved by https://github.com/microsoft/DeepSpeed/issues/4372
deepspeed >=0.8.2, <=0.9.3; platform_system != "Windows" and platform_system != "Darwin" # strict
deepspeed >=0.9.3, <=0.9.3; platform_system != "Windows" and platform_system != "Darwin" # strict
Loading