riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 02:02:21 +06:00
Code Issues Actions 33 Packages Projects Releases Wiki Activity

[Awq] Add peft support for AWQ (#28987)

Browse Source 
* add peft support for AWQ

* Update src/transformers/quantizers/quantizer_awq.py

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* fix

---------

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
This commit is contained in:
Younes Belkada 2024-02-19 01:31:39 +01:00 committed by GitHub
parent ce4fff0be7
commit 864c8e6ea3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/transformers/quantizers/quantizer_awq.py
View File

@ -11,8 +11,11 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import importlib.metadata
from typing import TYPE_CHECKING
from packaging import version
from .base import HfQuantizer
@ -105,6 +108,6 @@ class AwqQuantizer(HfQuantizer):
@property
def is_trainable(self):
# AWQ does not support neither QAT (Quantization Aware Training or PEFT yet.)
# TODO: if this is supported in the future, do a version check here.
return False
# AWQ supports PEFT fine-tuning from version 0.2.0
MIN_AWQ_VERSION_FOR_PEFT = "0.2.0"
return version.parse(importlib.metadata.version("autoawq")) >= version.parse(MIN_AWQ_VERSION_FOR_PEFT)