Open heyarny opened 1 year ago
Have u tried following these steps?
It seems to be something related to Pytorch
.
We are looking into updating jina
and docarray
versions in conda
Yes, according to the documentation my torch is using "mps" which is GPU supported.
Output of:
import torch
import math
# this ensures that the current MacOS version is at least 12.3+
print(torch.backends.mps.is_available())
# this ensures that the current current PyTorch installation was built with MPS activated.
print(torch.backends.mps.is_built())
True
True
This is my small setup:
flow.yml:
jtype: Flow
version: '1'
with:
port: 54321
executors:
- name: encoder
uses: 'jinahub://TransformerTorchEncoder/latest-gpu'
volumes: '.cache/huggingface:/root/.cache/huggingface'
install_requirements: true
- name: PrintExecutor
uses: print_executor/config.yml
Full output:
jina flow --uses flow.yml
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMWWWMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMWNNNNNNNWMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMNNNNNNNNNMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMWNNNNNNNNMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMWNNNWWMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMMMMMMMMMMMWxxxxxxxxxOMMMMMNxxxxxxxxx0MMMMMKddddddxkKWMMMMMMMMMMMMXOxdddxONMMMM
MMMMMMMMMMMMXllllllllldMMMMM0lllllllllxMMMMMOllllllllllo0MMMMMMMM0olllllllllo0MM
MMMMMMMMMMMMXllllllllldMMMMM0lllllllllxMMMMMOlllllllllllloWMMMMMdllllllllllllldM
MMMMMMMMMMMMXllllllllldMMMMM0lllllllllxMMMMMOllllllllllllloMMMM0lllllllllllllllK
MMMMMMMMMMMMKllllllllldMMMMM0lllllllllxMMMMMOllllllllllllllKMMM0lllllllllllllllO
MMMMMMMMMMMMKllllllllldMMMMM0lllllllllxMMMMMOllllllllllllll0MMMMollllllllllllllO
MWOkkkkk0MMMKlllllllllkMMMMM0lllllllllxMMMMMOllllllllllllll0MMMMMxlllllllllllllO
NkkkkkkkkkMMKlllllllloMMMMMM0lllllllllxMMMMMOllllllllllllll0MMMMMMWOdolllllllllO
KkkkkkkkkkNMKllllllldMMMMMMMMWWWWWWWWWMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MOkkkkkkk0MMKllllldXMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
MMWX00KXMMMMXxk0XMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
/opt/homebrew/bin/jina flow --uses flow.yml
╭──────────────┬──────────────────────────────────╮
│ Argument │ Value │
├──────────────┼──────────────────────────────────┤
│ cli │ flow │
│ env │ None │
│ inspect │ COLLECT │
│ log-config │ default │
│ name │ None │
│ quiet │ False │
│ quiet-error │ False │
│ uses │ flow.yml │
│ workspace │ None │
│ workspace-id │ 21f44c96c2de40f18d25158412a956d4 │
╰──────────────┴──────────────────────────────────╯
⠙ Installing dependencies from requirements.txt...Looking in links: https://download.pytorch.org/whl/torch_stable.html
⠴ Installing dependencies from requirements.txt...ERROR: Could not find a version that satisfies the requirement torch==1.9.0+cpu (from versions: 1.11.0, 1.12.0, 1.12.1, 1.13.0)
ERROR: No matching distribution found for torch==1.9.0+cpu
Traceback (most recent call last):
File "/opt/homebrew/bin/jina", line 8, in <module>
sys.exit(main())
File "/opt/homebrew/lib/python3.10/site-packages/jina_cli/__init__.py", line 143, in main
getattr(api, args.cli.replace('-', '_'))(args)
File "/opt/homebrew/lib/python3.10/site-packages/jina_cli/api.py", line 169, in flow
with f:
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/flow/base.py", line 1718, in __enter__
return self.start()
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/flow/builder.py", line 33, in arg_wrapper
return func(self, *args, **kwargs)
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/flow/base.py", line 1781, in start
self.enter_context(deployment)
File "/opt/homebrew/Cellar/python@3.10/3.10.8/Frameworks/Python.framework/Versions/3.10/lib/python3.10/contextlib.py", line 492, in enter_context
result = _cm_type.__enter__(cm)
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/deployments/__init__.py", line 121, in __enter__
return self.start()
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/deployments/__init__.py", line 668, in start
self.enter_context(self.shards[shard_id])
File "/opt/homebrew/Cellar/python@3.10/3.10.8/Frameworks/Python.framework/Versions/3.10/lib/python3.10/contextlib.py", line 492, in enter_context
result = _cm_type.__enter__(cm)
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/deployments/__init__.py", line 232, in __enter__
self._pods.append(PodFactory.build_pod(_args).start())
File "/opt/homebrew/lib/python3.10/site-packages/jina/orchestrate/pods/factory.py", line 36, in build_pod
cargs.uses = HubIO(_hub_args).pull()
File "/opt/homebrew/lib/python3.10/site-packages/hubble/executor/hubio.py", line 1362, in pull
install_package_dependencies(
File "/opt/homebrew/lib/python3.10/site-packages/hubble/executor/hubapi.py", line 209, in install_package_dependencies
install_requirements(requirements_file)
File "/opt/homebrew/lib/python3.10/site-packages/hubble/executor/helper.py", line 719, in install_requirements
subprocess.check_call(
File "/opt/homebrew/Cellar/python@3.10/3.10.8/Frameworks/Python.framework/Versions/3.10/lib/python3.10/subprocess.py", line 369, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/opt/homebrew/opt/python@3.10/bin/python3.10', '-m', 'pip', 'install', '--compile', '--default-timeout=1000', 'torch==1.9.0+cpu', 'transformers>=4.12.0', '-f', 'https://download.pytorch.org/whl/torch_stable.html']' returned non-zero exit status 1.
I think the requirements.txt needs to update the torch version. Otherwise, it would fail when using this executor.
@jpzhangvincent may u provide a PR with the required changes u consider?
I'm trying to get this to work on my macbook pro m1 max, but no matter what I do I get this error when trying to run it locally.
Anyone know how to fix this?