Skip to content

[GCP] Update tpu runtime version for doc #2602

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 1 commit into from
Sep 25, 2023
Merged

[GCP] Update tpu runtime version for doc #2602

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 4 additions & 4 deletions docs/source/reference/tpu.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,7 +128,7 @@ To use a TPU Node, set the following in a task YAML's ``resources`` field:
instance_type: n1-highmem-8
accelerators: tpu-v2-8
accelerator_args:
runtime_version: 2.5.0 # optional, TPU runtime version.
runtime_version: 2.12.0 # optional, TPU runtime version.

The above YAML considers :code:`n1-highmem-8` as the host machine and :code:`tpu-v2-8` as the TPU node resource.
You can modify the host instance type or the TPU type.
Expand All @@ -143,7 +143,7 @@ Here is a complete task YAML that runs `MNIST training <https://cloud.google.com
resources:
accelerators: tpu-v2-8
accelerator_args:
runtime_version: 2.5.0 # optional, TPU runtime version.
runtime_version: 2.12.0 # optional, TPU runtime version.

# TPU node requires loading data from a GCS bucket.
# We use SkyPilot Storage to mount a GCS bucket to /dataset.
Expand All @@ -162,7 +162,7 @@ Here is a complete task YAML that runs `MNIST training <https://cloud.google.com
else
conda create -n mnist python=3.8 -y
conda activate mnist
pip install tensorflow==2.5.0 tensorflow-datasets tensorflow-model-optimization cloud-tpu-client
pip install tensorflow==2.12.0 tensorflow-datasets tensorflow-model-optimization cloud-tpu-client
fi

run: |
Expand Down Expand Up @@ -306,4 +306,4 @@ To submit more jobs to the same TPU Pod, use :code:`sky exec`:

.. code-block:: console

$ sky exec mycluster examples/tpu/cifar_pod.yaml
$ sky exec mycluster examples/tpu/cifar_pod.yaml
4 changes: 2 additions & 2 deletions docs/source/reference/yaml-spec.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,8 +117,8 @@ Available fields:
# requested and should work for either case. If passing in an incompatible
# version, GCP will throw an error during provisioning.
accelerator_args:
# Default is "2.5.0" for TPU node and "tpu-vm-base" for TPU VM.
runtime_version: 2.5.0
# Default is "2.12.0" for TPU node and "tpu-vm-base" for TPU VM.
runtime_version: 2.12.0
tpu_name: mytpu
tpu_vm: False # False to use TPU nodes (the default); True to use TPU VMs.

Expand Down