Merged
Conversation
jfrery
force-pushed
the
chore/add_llama_finetuning_benchmark
branch
from
abb4b2e to
b04bad6
Compare
jfrery
force-pushed
the
chore/add_llama_finetuning_benchmark
branch
from
6595bad to
188f773
Compare
jfrery
force-pushed
the
chore/add_llama_finetuning_benchmark
branch
from
3018276 to
28c1410
Compare
jfrery
force-pushed
the
chore/add_llama_finetuning_benchmark
branch
from
ba90eb8 to
147c7fd
Compare
kcelia
reviewed
Jun 6, 2025
kcelia
reviewed
Jun 6, 2025
|
|
||
| # Install PyTorch with appropriate CUDA support | ||
| if [ "${{ matrix.device }}" == "gpu" ]; then | ||
| pip install torch>=2.0.0+cu118 --extra-index-url https://download.pytorch.org/whl/cu118 |
Contributor
There was a problem hiding this comment.
on this machine 'Nvidia -smi' already works, you just need to enable Concrete GPU:
- name: Install GPU concrete-python
id: install-gpu-concrete-python
run: |
poetry run pip show concrete-python || echo "concrete-python not installed"
CONCRETE_WITH_VERSION=$(poetry run pip freeze | grep concrete-python)
poetry run pip uninstall -y concrete-python
poetry run pip install --extra-index-url https://pypi.zama.ai/gpu $CONCRETE_WITH_VERSION
poetry run pip show concrete-python || echo "concrete-python not installed"
kcelia
reviewed
Jun 6, 2025
| - name: Set up Python | ||
| uses: actions/setup-python@42375524e23c412d93fb67b49958b491fce71c38 | ||
| with: | ||
| python-version: "3.10" |
Contributor
There was a problem hiding this comment.
if we follow the logic of our CI, you should pick p3.12
kcelia
reviewed
Jun 6, 2025
|
|
||
| # Install any additional requirements from the project | ||
| if [ -f "requirements.txt" ]; then | ||
| pip install -r requirements.txt |
kcelia
reviewed
Jun 6, 2025
| fi | ||
|
|
||
| # Install accelerate BEFORE other dependencies to ensure correct version | ||
| pip install 'accelerate>=1.1.0' |
Contributor
There was a problem hiding this comment.
why you don't gather all the package dependancies in one place ?
kcelia
reviewed
Jun 6, 2025
| info["swap"] = get_size(psutil.swap_memory().total) | ||
|
|
||
| # Check for GPU information | ||
| if torch.cuda.is_available(): |
Contributor
There was a problem hiding this comment.
if the model runs on GPU, you should check concrete gpu as well no?
kcelia
reviewed
Jun 6, 2025
| if force_cpu or device_type == "cpu": | ||
| return "cpu" | ||
|
|
||
| if device_type == "gpu": |
Contributor
There was a problem hiding this comment.
add a check for concrete gpu ?
Coverage passed ✅Coverage details
|
|
kcelia
approved these changes
Oct 2, 2025
Contributor
kcelia
left a comment
kcelia
left a comment
There was a problem hiding this comment.
In my opinion, it's better to use the pre-installed cuda of the image and just install concrete gpu.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.