ML CI: Linux aarch64 (#39666)
* ML CI: Linux aarch64 * Add config files * No aarch64 tag * Don't specify image * Use amazonlinux image Co-authored-by: kwryankrattiger <80296582+kwryankrattiger@users.noreply.github.com> * Update and require * GCC is too old * Fix some builds * xgboost doesn't support old GCC + cuda * Run on newer Ubuntu * Remove mxnet * Try aarch64 range * Use main branch * Conflict applies to all targets * cuda only required when +cuda * Use tagged version * Comment out tf-estimator * Add ROCm, use newer Ubuntu * Remove ROCm --------- Co-authored-by: kwryankrattiger <80296582+kwryankrattiger@users.noreply.github.com>
This commit is contained in:
parent
e83536de38
commit
32ce278a51
@ -735,6 +735,52 @@ ml-linux-x86_64-rocm-build:
|
||||
- artifacts: True
|
||||
job: ml-linux-x86_64-rocm-generate
|
||||
|
||||
########################################
|
||||
# Machine Learning - Linux aarch64 (CPU)
|
||||
########################################
|
||||
.ml-linux-aarch64-cpu:
|
||||
extends: [ ".linux_aarch64" ]
|
||||
variables:
|
||||
SPACK_CI_STACK_NAME: ml-linux-aarch64-cpu
|
||||
|
||||
ml-linux-aarch64-cpu-generate:
|
||||
extends: [ ".generate-aarch64", .ml-linux-aarch64-cpu ]
|
||||
image: ghcr.io/spack/ubuntu-24.04:v2024-09-05-v2
|
||||
|
||||
ml-linux-aarch64-cpu-build:
|
||||
extends: [ ".build", ".ml-linux-aarch64-cpu" ]
|
||||
trigger:
|
||||
include:
|
||||
- artifact: jobs_scratch_dir/cloud-ci-pipeline.yml
|
||||
job: ml-linux-aarch64-cpu-generate
|
||||
strategy: depend
|
||||
needs:
|
||||
- artifacts: True
|
||||
job: ml-linux-aarch64-cpu-generate
|
||||
|
||||
#########################################
|
||||
# Machine Learning - Linux aarch64 (CUDA)
|
||||
#########################################
|
||||
.ml-linux-aarch64-cuda:
|
||||
extends: [ ".linux_aarch64" ]
|
||||
variables:
|
||||
SPACK_CI_STACK_NAME: ml-linux-aarch64-cuda
|
||||
|
||||
ml-linux-aarch64-cuda-generate:
|
||||
extends: [ ".generate-aarch64", .ml-linux-aarch64-cuda ]
|
||||
image: ghcr.io/spack/ubuntu-24.04:v2024-09-05-v2
|
||||
|
||||
ml-linux-aarch64-cuda-build:
|
||||
extends: [ ".build", ".ml-linux-aarch64-cuda" ]
|
||||
trigger:
|
||||
include:
|
||||
- artifact: jobs_scratch_dir/cloud-ci-pipeline.yml
|
||||
job: ml-linux-aarch64-cuda-generate
|
||||
strategy: depend
|
||||
needs:
|
||||
- artifacts: True
|
||||
job: ml-linux-aarch64-cuda-generate
|
||||
|
||||
#########################################
|
||||
# Machine Learning - Darwin aarch64 (MPS)
|
||||
#########################################
|
||||
|
@ -0,0 +1,85 @@
|
||||
spack:
|
||||
view: false
|
||||
packages:
|
||||
all:
|
||||
require:
|
||||
- target=aarch64
|
||||
- ~cuda
|
||||
- ~rocm
|
||||
mpi:
|
||||
require: openmpi
|
||||
|
||||
specs:
|
||||
# Horovod
|
||||
- py-horovod
|
||||
|
||||
# Hugging Face
|
||||
- py-transformers
|
||||
|
||||
# JAX
|
||||
- py-jax
|
||||
- py-jaxlib
|
||||
|
||||
# Keras
|
||||
- py-keras backend=tensorflow
|
||||
- py-keras backend=jax
|
||||
- py-keras backend=torch
|
||||
- py-keras-applications
|
||||
- py-keras-preprocessing
|
||||
- py-keras2onnx
|
||||
|
||||
# PyTorch
|
||||
- py-botorch
|
||||
- py-efficientnet-pytorch
|
||||
- py-gpytorch
|
||||
- py-kornia
|
||||
- py-lightning
|
||||
- py-pytorch-gradual-warmup-lr
|
||||
- py-pytorch-lightning
|
||||
- py-segmentation-models-pytorch
|
||||
- py-timm
|
||||
- py-torch
|
||||
- py-torch-cluster
|
||||
- py-torch-geometric
|
||||
- py-torch-nvidia-apex
|
||||
- py-torch-scatter
|
||||
- py-torch-sparse
|
||||
- py-torch-spline-conv
|
||||
- py-torchaudio
|
||||
- py-torchdata
|
||||
- py-torchfile
|
||||
- py-torchgeo
|
||||
- py-torchmetrics
|
||||
- py-torchtext
|
||||
- py-torchvision
|
||||
- py-vector-quantize-pytorch
|
||||
|
||||
# scikit-learn
|
||||
- py-scikit-learn
|
||||
- py-scikit-learn-extra
|
||||
|
||||
# TensorBoard
|
||||
- py-tensorboard
|
||||
- py-tensorboard-data-server
|
||||
- py-tensorboard-plugin-wit
|
||||
- py-tensorboardx
|
||||
|
||||
# TensorFlow
|
||||
- py-tensorflow
|
||||
- py-tensorflow-datasets
|
||||
- py-tensorflow-hub
|
||||
- py-tensorflow-metadata
|
||||
- py-tensorflow-probability
|
||||
|
||||
# XGBoost
|
||||
- py-xgboost
|
||||
|
||||
ci:
|
||||
pipeline-gen:
|
||||
- build-job:
|
||||
image:
|
||||
name: ghcr.io/spack/ubuntu-24.04:v2024-09-05-v2
|
||||
entrypoint: ['']
|
||||
|
||||
cdash:
|
||||
build-group: Machine Learning
|
@ -0,0 +1,91 @@
|
||||
spack:
|
||||
view: false
|
||||
packages:
|
||||
all:
|
||||
require:
|
||||
- target=aarch64
|
||||
- ~rocm
|
||||
- +cuda
|
||||
- cuda_arch=80
|
||||
llvm:
|
||||
# https://github.com/spack/spack/issues/27999
|
||||
require: ~cuda
|
||||
mpi:
|
||||
require: openmpi
|
||||
|
||||
specs:
|
||||
# Horovod
|
||||
- py-horovod
|
||||
|
||||
# Hugging Face
|
||||
- py-transformers
|
||||
|
||||
# JAX
|
||||
- py-jax
|
||||
- py-jaxlib
|
||||
|
||||
# Keras
|
||||
- py-keras backend=tensorflow
|
||||
- py-keras backend=jax
|
||||
- py-keras backend=torch
|
||||
- py-keras-applications
|
||||
- py-keras-preprocessing
|
||||
- py-keras2onnx
|
||||
|
||||
# PyTorch
|
||||
- py-botorch
|
||||
- py-efficientnet-pytorch
|
||||
- py-gpytorch
|
||||
- py-kornia
|
||||
- py-lightning
|
||||
- py-pytorch-gradual-warmup-lr
|
||||
- py-pytorch-lightning
|
||||
- py-segmentation-models-pytorch
|
||||
- py-timm
|
||||
- py-torch
|
||||
- py-torch-cluster
|
||||
- py-torch-geometric
|
||||
- py-torch-nvidia-apex
|
||||
- py-torch-scatter
|
||||
- py-torch-sparse
|
||||
- py-torch-spline-conv
|
||||
- py-torchaudio
|
||||
- py-torchdata
|
||||
- py-torchfile
|
||||
- py-torchgeo
|
||||
- py-torchmetrics
|
||||
# torchtext requires older pytorch, which requires older cuda, which doesn't support newer GCC
|
||||
# - py-torchtext
|
||||
- py-torchvision
|
||||
- py-vector-quantize-pytorch
|
||||
|
||||
# scikit-learn
|
||||
- py-scikit-learn
|
||||
- py-scikit-learn-extra
|
||||
|
||||
# TensorBoard
|
||||
- py-tensorboard
|
||||
- py-tensorboard-data-server
|
||||
- py-tensorboard-plugin-wit
|
||||
- py-tensorboardx
|
||||
|
||||
# TensorFlow
|
||||
- py-tensorflow
|
||||
- py-tensorflow-datasets
|
||||
- py-tensorflow-hub
|
||||
- py-tensorflow-metadata
|
||||
- py-tensorflow-probability
|
||||
|
||||
# XGBoost
|
||||
# xgboost requires older cuda, which doesn't support newer GCC
|
||||
# - py-xgboost
|
||||
|
||||
ci:
|
||||
pipeline-gen:
|
||||
- build-job:
|
||||
image:
|
||||
name: ghcr.io/spack/ubuntu-24.04:v2024-09-05-v2
|
||||
entrypoint: ['']
|
||||
|
||||
cdash:
|
||||
build-group: Machine Learning
|
Loading…
Reference in New Issue
Block a user