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 23 Packages Projects Releases Wiki Activity

Expose offload_buffers parameter of accelerate to PreTrainedModel.from_pretrained method (#28755)

Browse Source 
Expose offload_buffers parameter to from_pretrained method
This commit is contained in:
Song Fuchang 2024-03-01 09:12:51 +08:00 committed by GitHub
parent 0ad770c373
commit 5ee0868a4b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/transformers/modeling_utils.py
View File

@ -2745,6 +2745,8 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
If `True`, will temporarily offload the CPU state dict to the hard drive to avoid getting out of CPU
RAM if the weight of the CPU state dict + the biggest shard of the checkpoint does not fit. Defaults to
`True` when there is some disk offload.
offload_buffers (`bool`, *optional*):
Whether or not to offload the buffers with the model parameters.
quantization_config (`Union[QuantizationConfigMixin,Dict]`, *optional*):
A dictionary of configuration parameters or a QuantizationConfigMixin object for quantization (e.g
bitsandbytes, gptq). There may be other quantization-related kwargs, including `load_in_4bit` and
@ -2835,6 +2837,7 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
max_memory = kwargs.pop("max_memory", None)
offload_folder = kwargs.pop("offload_folder", None)
offload_state_dict = kwargs.pop("offload_state_dict", False)
offload_buffers = kwargs.pop("offload_buffers", False)
load_in_8bit = kwargs.pop("load_in_8bit", False)
load_in_4bit = kwargs.pop("load_in_4bit", False)
quantization_config = kwargs.pop("quantization_config", None)
@ -3554,6 +3557,7 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
"device_map": device_map,
"offload_dir": offload_folder,
"offload_index": offload_index,
"offload_buffers": offload_buffers,
}
if "skip_keys" in inspect.signature(dispatch_model).parameters:
device_map_kwargs["skip_keys"] = model._skip_keys_device_placement