github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-27 12:43:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

[npu] add npu support for gemini and zero (#5067)

Browse Source 
* [npu] setup device utils (#5047)

* [npu] add npu device support

* [npu] support low level zero

* [test] update npu zero plugin test

* [hotfix] fix import

* [test] recover tests

* [npu] gemini support npu (#5052)

* [npu] refactor device utils

* [gemini] support npu

* [example] llama2+gemini support npu

* [kernel] add arm cpu adam kernel (#5065)

* [kernel] add arm cpu adam

* [optim] update adam optimizer

* [kernel] arm cpu adam remove bf16 support
This commit is contained in:
Hongxin Liu
2023-11-20 16:12:41 +08:00
committed by GitHub
parent 8d56c9c389
commit e5ce4c8ea6
46 changed files with 994 additions and 233 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
colossalai/amp/naive_amp/grad_scaler/dynamic_grad_scaler.py
View File

@@ -5,6 +5,8 @@ from typing import Optional
import torch
from colossalai.utils.device import get_current_device
from .base_grad_scaler import BaseGradScaler
__all__ = ["DynamicGradScaler"]
@@ -37,12 +39,12 @@ class DynamicGradScaler(BaseGradScaler):
):
super().__init__(initial_scale, verbose)
if min_scale:
self._min_scale = torch.cuda.FloatTensor([min_scale])
self._min_scale = torch.tensor([min_scale], device=get_current_device(), dtype=torch.float)
else:
self._min_scale = None
if max_scale:
self._max_scale = torch.cuda.FloatTensor([max_scale])
self._max_scale = torch.tensor([max_scale], device=get_current_device(), dtype=torch.float)
else:
self._max_scale = None
@@ -115,7 +117,7 @@ class DynamicGradScaler(BaseGradScaler):
return state_dict
def load_state_dict(self, state_dict):
self._scale = state_dict["scale"].cuda(torch.cuda.current_device())
self._scale = state_dict["scale"].to(get_current_device())
self._growth_factor = state_dict["growth_factor"]
self._backoff_factor = state_dict["backoff_factor"]
self._hysteresis = state_dict["hysteresis"]