-
Notifications
You must be signed in to change notification settings - Fork 215
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.
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
Llm phi #979
base: develop
Are you sure you want to change the base?
Llm phi #979
Conversation
@porteratzo can you take care of failed pipelines and ask for review agian? |
6d82206
to
9b68b7f
Compare
"outputs": [], | ||
"source": [ | ||
"!pip install git+https://github.com/intel/openfl.git\n", | ||
"!cd /home/oamontoy/workspace/openfl/openfl-tutorials/experimental/Phi3 && pip install -r ../requirements_workflow_interface.txt\n", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can you remove the absolute path and. keep general like
!pip install git+https://github.com/securefederatedai/openfl.git
!pip install -r workflow_interface_requirements.txt
"!pip install git+https://github.com/intel/openfl.git\n", | ||
"!cd /home/oamontoy/workspace/openfl/openfl-tutorials/experimental/Phi3 && pip install -r ../requirements_workflow_interface.txt\n", | ||
"!pip install numpy --upgrade\n", | ||
"!pip install transformers peft datasets trl" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
when running "!pip install datasets trl" getting the below error
-- Running cmake for PyArrow
cmake -DCMAKE_INSTALL_PREFIX=/private/var/folders/bj/xlf52n2s7zx24gvllky2f_wh0000gn/T/pip-install-55o676u7/pyarrow_50ccfaecb1b946ff8ad649bdf24d7546/build/lib.macosx-10.9-x86_64-cpython-311/pyarrow -DPYTHON_EXECUTABLE=/Users/rkrishne/anaconda3/bin/python -DPython3_EXECUTABLE=/Users/rkrishne/anaconda3/bin/python -DPYARROW_CXXFLAGS= -DPYARROW_BUNDLE_ARROW_CPP=off -DPYARROW_BUNDLE_CYTHON_CPP=off -DPYARROW_GENERATE_COVERAGE=off -DCMAKE_BUILD_TYPE=release /private/var/folders/bj/xlf52n2s7zx24gvllky2f_wh0000gn/T/pip-install-55o676u7/pyarrow_50ccfaecb1b946ff8ad649bdf24d7546
error: command 'cmake' failed: No such file or directory
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for pyarrow
Failed to build pyarrow
ERROR: Could not build wheels for pyarrow, which is required to install pyproject.toml-based projects
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
to resolve tat error we need to install cmake package but even after installing cmake package getting the below error
/private/var/folders/bj/xlf52n2s7zx24gvllky2f_wh0000gn/T/pip-install-doldjp2l/pyarrow_53bf7158da274fff8d32761768c8e19b/pyarrow/src/arrow/python/arrow_to_pandas.cc:208:16:` error: no member named 'LARGE_LIST_VIEW' in 'arrow::Type'; did you mean 'LARGE_LIST'?
case Type::LARGE_LIST_VIEW: {
~~~~~~^~~~~~~~~~~~~~~
LARGE_LIST
/anaconda3/include/arrow/type_fwd.h:410:5: note: 'LARGE_LIST' declared here
LARGE_LIST = 36,
^
fatal error: too many errors emitted, stopping now [-ferror-limit=]
20 errors generated.
make[2]: *** [CMakeFiles/arrow_python.dir/pyarrow/src/arrow/python/arrow_to_pandas.cc.o] Error 1
make[1]: *** [CMakeFiles/arrow_python.dir/all] Error 2
make: *** [all] Error 2
error: command '/usr/local/bin/cmake' failed with exit code 2
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for pyarrow
Failed to build pyarrow
ERROR: Could not build wheels for pyarrow, which is required to install pyproject.toml-based projects
Note: you may need to restart the kernel to use updated packages.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am unable to replicate the error, tested with a fresh python3.10 env using conda, also used --no-cache to make sure packages where being downloaded
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I tired with python 3.10 version and was able to install without any error. But when i tired with python 3.11 version then we get the above issue.
We have to specify somewhere as a prerequisites that it only works with python 3.10.
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
* Replace xml.etree.ElementTree.parse with its defusedxml Signed-off-by: Chaurasiya, Payal <[email protected]> * convert to json Signed-off-by: Chaurasiya, Payal <[email protected]> * Fix memory logs and create pdf Signed-off-by: Chaurasiya, Payal <[email protected]> --------- Signed-off-by: Chaurasiya, Payal <[email protected]> Signed-off-by: porteratzo <[email protected]>
) * Create openfl-docker-build.yml * remove test event Signed-off-by: porteratzo <[email protected]>
Signed-off-by: noopur <[email protected]> Signed-off-by: porteratzo <[email protected]>
Bumps [pytest-asyncio](https://github.com/pytest-dev/pytest-asyncio) from 0.25.0 to 0.25.1. - [Release notes](https://github.com/pytest-dev/pytest-asyncio/releases) - [Commits](pytest-dev/pytest-asyncio@v0.25.0...v0.25.1) --- updated-dependencies: - dependency-name: pytest-asyncio dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Signed-off-by: porteratzo <[email protected]>
Signed-off-by: porteratzo <[email protected]>
@psfoley should we direct this PR to openFL-contrib instead? |
A jupyter notebook to demo finetuning with the Phi3 LLM model. https://huggingface.co/docs/transformers/main/en/model_doc/phi3
We use the math_10k dataset https://github.com/AGI-Edgerunners/LLM-Adapters/tree/main
We download the dataset from github and check its hash and preprocess the dataset.
We load the phi model and prepare it for PEFT training.
Then we create a workflow that extracts the peft parameters and excludes sending the full model to the collaborators, the collaborators reconstruct the model from the original model and only modify the peft parameters. And after training and evaluation return the peft parameters again to be used with fed average.