Add tutorial notebook for PEFT #11216
Conversation
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
|
Documentation preview for fb7f04d will be available when this CircleCI job completes successfully. More info
|
github-actions
bot
added
area/docs
B-Step62
changed the title
mlflow-automation
requested review from
BenWilson2,
daniellok-db,
harupy,
serena-ruan and
WeichenXu123
and removed request for
mlflow-automation
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
| "cell_type": "raw", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "<p></p>\n", |
There was a problem hiding this comment.
Found that we can insert the download link in the notebook using "raw" cell, so we can remove the workaround of putting the button in index page. If this sounds good, I will apply same to the other Transformers notebook in the follow-up cc: @BenWilson2
There was a problem hiding this comment.
The original reason for not putting the download link in the notebook itself was purely so that users wouldn't be confused with a download button embedded in their local copy of the notebook and, if they clicked on it, having a bunch of stateless copies of the notebook fill up their downloads folder.
That might be a bit overkill as a cautionary confusion prevention step, though. I'm fine with embedding it :)
There was a problem hiding this comment.
Ooh sorry didn't remember that was the reason but just thought it was nbsphinx thing...
Let me stick to embedded button here for simplify the top page (as Transformers now have more notebooks so it might be a bit hard to find the corresponding link from the list), but keep the other flavors pages as they are.
There was a problem hiding this comment.
No, I agree with you. Let's file a ticket for next sprint to migrate to the raw embed within notebook approach. It's cleaner. We should do this for all notebooks in all sections. Probably a 0.5 point ticket :)
docs/source/llms/transformers/tutorials/fine-tuning/transformers-peft.ipynb
Outdated
Show resolved
Hide resolved
| } | ||
| ], | ||
| "source": [ | ||
| "% nvidia-smi" |
There was a problem hiding this comment.
Does Jupyter support this magic command?
There was a problem hiding this comment.
What happens if I run-all this notebook and nvidia-smi is not installed?
There was a problem hiding this comment.
Does Jupyter support this magic command?
Nope, fixed to %sh nvidia-smi.
What happens if I run-all this notebook and nvidia-smi is not installed?
The first cell should fail with command not found, which is desired as we don't expect users to run this notebook without CUDA GPU backend.
B-Step62
force-pushed
the
peft-notebook
branch
from
d67dd52
to
fb7f04d
Compare
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
Signed-off-by: B-Step62 <yuki.watanabe@databricks.com>
馃洜 DevTools 馃洜
Install mlflow from this PR
Checkout with GitHub CLI
Related Issues/PRs
#xxxWhat changes are proposed in this pull request?
Add a tutorial notebook for PEFT (QLoRA) with MLflow. The main docs will be updated in a separate PR.
Note: It seems the new linter confuses the shell script cell with Python.
How is this PR tested?
Does this PR require documentation update?
Release Notes
Is this a user-facing change?
What component(s), interfaces, languages, and integrations does this PR affect?
Components
area/artifacts: Artifact stores and artifact loggingarea/build: Build and test infrastructure for MLflowarea/deployments: MLflow Deployments client APIs, server, and third-party Deployments integrationsarea/docs: MLflow documentation pagesarea/examples: Example codearea/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registryarea/models: MLmodel format, model serialization/deserialization, flavorsarea/recipes: Recipes, Recipe APIs, Recipe configs, Recipe Templatesarea/projects: MLproject format, project running backendsarea/scoring: MLflow Model server, model deployment tools, Spark UDFsarea/server-infra: MLflow Tracking server backendarea/tracking: Tracking Service, tracking client APIs, autologgingInterface
area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev serverarea/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Modelsarea/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registryarea/windows: Windows supportLanguage
language/r: R APIs and clientslanguage/java: Java APIs and clientslanguage/new: Proposals for new client languagesIntegrations
integrations/azure: Azure and Azure ML integrationsintegrations/sagemaker: SageMaker integrationsintegrations/databricks: Databricks integrationsHow should the PR be classified in the release notes? Choose one:
rn/none- No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" sectionrn/breaking-change- The PR will be mentioned in the "Breaking Changes" sectionrn/feature- A new user-facing feature worth mentioning in the release notesrn/bug-fix- A user-facing bug fix worth mentioning in the release notesrn/documentation- A user-facing documentation change worth mentioning in the release notes