github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-10 05:20:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

[npu] change device to accelerator api (#5239)

Browse Source 
* update accelerator

* fix timer

* fix amp

* update

* fix

* update bug

* add error raise

* fix autocast

* fix set device

* remove doc accelerator

* update doc

* update doc

* update doc

* use nullcontext

* update cpu

* update null context

* change time limit for example

* udpate

* update

* update

* update

* [npu] polish accelerator code

---------

Co-authored-by: Xuanlei Zhao <xuanlei.zhao@gmail.com>
Co-authored-by: zxl <43881818+oahzxl@users.noreply.github.com>
This commit is contained in:
Hongxin Liu
2024-01-09 10:20:05 +08:00
committed by GitHub
parent dd2c28a323
commit d202cc28c0
128 changed files with 1773 additions and 868 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
colossalai/auto_parallel/offload/solver.py
View File

@@ -11,7 +11,7 @@ except:
import torch
from torch.fx.node import Node
from colossalai.utils.device import get_current_device
from colossalai.accelerator import get_accelerator
from .region import Region
from .training_simulator import AsynTrainingSimulator, SynTrainingSimulator, TrainingSimulator
@@ -57,7 +57,10 @@ class Solver(ABC):
if memory_budget > 0:
self.memory_budget = memory_budget * self.error_factor
else:
self.memory_budget = torch.cuda.get_device_properties(get_current_device()).total_memory * self.error_factor
self.memory_budget = (
torch.cuda.get_device_properties(get_accelerator().get_current_device()).total_memory
* self.error_factor
)
self.link_to_bandwidth: Dict[str, Dict[float, float]] = self._profile_bandwidth()
self.comp_power: float = self._extract_computing_power()