Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[OV]: Fixed inference after 4 bit weight compression #569

Merged
merged 5 commits into from
Feb 21, 2024
Merged

[OV]: Fixed inference after 4 bit weight compression #569

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
6dc4524
[OV]: Fixed inferece after 4 bit weight compression
AlexKoff88 Feb 19, 2024
7188696
Fixed issue
AlexKoff88 Feb 19, 2024
d06a344
Update optimum/intel/openvino/modeling_decoder.py
AlexKoff88 Feb 20, 2024
c816e96
Applied comments
AlexKoff88 Feb 20, 2024
ede353e
Fixed issue when request is None
AlexKoff88 Feb 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions optimum/intel/openvino/modeling_decoder.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -419,7 +419,8 @@ def prepare_inputs(
# past_key_values are not used explicitly, instead they are handled inside the model
if past_key_values is None:
# This is the first iteration in a sequence, reset all states
self.request.reset_state()
if self.request is not None:
self.request.reset_state()
# Set initial value for the next beam_idx input that will be used at the current iteration
# and will be optionally updated by _reorder_cache at the next iterations if beam_search is used
self.next_beam_idx = np.arange(batch_size, dtype=int)
Expand Down Expand Up @@ -592,7 +593,10 @@ def _from_pretrained(
else:
init_cls = cls

causal_model = init_cls(model=model, config=config, model_save_dir=model_cache_path.parent, **kwargs)
enable_compilation = kwargs.pop("compile", True) and not load_in_4bit
causal_model = init_cls(
model=model, config=config, model_save_dir=model_cache_path.parent, compile=enable_compilation, **kwargs
)

if load_in_4bit:
if not is_nncf_available():
Expand Down
Loading