Modifications to "get_working_device" function in Pytorch. Added the capability to set a Pytorch device using the "set_device" function #891
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
lior-dikstein
requested review from
reuvenperetz
and removed request for
ofirgo
reuvenperetz
requested changes
Dec 28, 2023
| @@ -25,4 +25,4 @@ | |||
| from model_compression_toolkit.core.common.mixed_precision.kpi_tools.kpi import KPI | |||
| from model_compression_toolkit.core.common.mixed_precision.mixed_precision_quantization_config import MixedPrecisionQuantizationConfig, MixedPrecisionQuantizationConfigV2 | |||
| from model_compression_toolkit.core.keras.kpi_data_facade import keras_kpi_data, keras_kpi_data_experimental | |||
| from model_compression_toolkit.core.pytorch.kpi_data_facade import pytorch_kpi_data, pytorch_kpi_data_experimental | |||
| from model_compression_toolkit.core.pytorch.kpi_data_facade import pytorch_kpi_data, pytorch_kpi_data_experimental | |||
There was a problem hiding this comment.
Remove unneeded changes
| Args: | ||
| device_name (str): The name of the device, e.g., 'cuda:0' or 'cpu'. | ||
|
|
||
| If the specified device is not valid or available, it prints an error message without changing the current device. |
There was a problem hiding this comment.
Do you think we should alert or raise an error? If just a warning - fix the comment.
There was a problem hiding this comment.
I think a warning. Changed the comment.
| else: | ||
| logger.Logger.warning(message) | ||
|
|
||
| def get_device(self): |
| return self.DEVICE | ||
|
|
||
| @staticmethod | ||
| def is_valid_device(device_name: str): |
There was a problem hiding this comment.
returned type is missing
|
|
||
| return True, "Valid device" | ||
|
|
||
| return True, "Valid device" |
There was a problem hiding this comment.
What if it's 'cpu'? I think it should be checked here before True is returned
There was a problem hiding this comment.
There isn't a check for CPU because it is always a valid choice. This checks that if you choose a GPU it exists.
There was a problem hiding this comment.
I think it's better to verify that it's 'cpu' and not unrelated string, but it's up to you
| device_manager = DeviceManager() | ||
| device_manager.set_device(device_name) | ||
|
|
||
| def get_working_device(): |
added 4 commits
December 31, 2023 12:07
…capability to set a Pytorch device using the "set_device" function. Implemented using DeviceManager with Singleton Pattern.
…capability to set a Pytorch device using the "set_device" function. Implemented using DeviceManager with Singleton Pattern.
lior-dikstein
force-pushed
the
pytorch_device
branch
from
9c73d22 to
7c848d3
Compare
reuvenperetz
approved these changes
Dec 31, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Pull Request Description:
Refine PyTorch device management using DeviceManager singleton class.
DeviceManagerfor consistent and centralized management of PyTorch devices.get_working_devicefunction in PyTorch for improved retrieval of the current device setting using the singletonDeviceManager.set_devicefunction usingDeviceManager, enabling the capability to set the PyTorch device across various project parts dynamically.DeviceManagerto handle specific CUDA device indices ('cuda:x') and the default CUDA device ('cuda').Checklist before requesting a review: