ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

[LLM Doc] Add optimize_model doc in transformers api (#8957)

Browse source 
* add optimize in from_pretrained

* add api doc for load_low_bit

* update api docs following comments

* update api docs

* update

* reord comments
This commit is contained in:
SONG Ge 2023-09-13 10:42:33 +08:00 committed by GitHub
parent c32c260ce2
commit 7132ef6081
1 changed files with 15 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
python/llm/src/bigdl/llm/transformers/model.py
View file

@ -54,13 +54,18 @@ class _BaseAutoModelClass:
Load a model from a directory or the HF Hub. Use load_in_4bit or load_in_low_bit parameter Load a model from a directory or the HF Hub. Use load_in_4bit or load_in_low_bit parameter
the weight of model's linears can be loaded to low-bit format, like int4, int5 and int8. the weight of model's linears can be loaded to low-bit format, like int4, int5 and int8.
Two new arguments are added to extend Hugging Face's from_pretrained method as follows: Three new arguments are added to extend Hugging Face's from_pretrained method as follows:
:param load_in_4bit: boolean value, True means load linear's weight to symmetric int 4. :param load_in_4bit: boolean value, True means load linear's weight to symmetric int 4.
Default to be False.
:param load_in_low_bit: str value, options are sym_int4, asym_int4, sym_int5, asym_int5 :param load_in_low_bit: str value, options are sym_int4, asym_int4, sym_int5, asym_int5
or sym_int8. sym_int4 means symmetric int 4, asym_int4 means or sym_int8. sym_int4 means symmetric int 4, asym_int4 means
asymmetric int 4, etc. Relevant low bit optimizations will asymmetric int 4, etc. Relevant low bit optimizations will
be applied to the model. be applied to the model.
:param optimize_model: boolean value, Whether to further optimize the low_bit llm model.
Default to be True.
:return: a model instance
""" """
pretrained_model_name_or_path = kwargs.get("pretrained_model_name_or_path", None) \ pretrained_model_name_or_path = kwargs.get("pretrained_model_name_or_path", None) \
if len(args) == 0 else args[0] if len(args) == 0 else args[0]
@ -128,6 +133,15 @@ class _BaseAutoModelClass:
pretrained_model_name_or_path, pretrained_model_name_or_path,
*model_args, *model_args,
**kwargs): **kwargs):
"""
Load a low bit optimized model (including INT4, INT5 and INT8) from a saved ckpt.
:param pretrained_model_name_or_path: str value, Path to load the optimized model ckpt.
:param optimize_model: boolean value, Whether to further optimize the low_bit llm model.
Default to be True.
:return: a model instance
"""
from transformers.modeling_utils import no_init_weights, get_state_dict_dtype from transformers.modeling_utils import no_init_weights, get_state_dict_dtype
from transformers.dynamic_module_utils import resolve_trust_remote_code, \ from transformers.dynamic_module_utils import resolve_trust_remote_code, \
get_class_from_dynamic_module get_class_from_dynamic_module