Skip to content

Fix: Skip weight initialization for quantized int8 models #42686

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

Closed
AnjaliiD wants to merge 1 commit into from
Closed

Fix: Skip weight initialization for quantized int8 models #42686

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
5 changes: 5 additions & 0 deletions src/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4116,6 +4116,11 @@ def _load_pretrained_model(
# This is not true but for now we assume only best-case scenario with deepspeed, i.e. perfectly matching checkpoints
missing_keys, unexpected_keys, mismatched_keys, misc = set(), set(), set(), set()
else:
# Skip weight initialization for quantized models (int8 weights)
if is_quantized:
print("Skipping weight initialization for quantized model.")
return missing_keys, unexpected_keys, mismatched_keys, None, set()

all_pointer = set()
# Checkpoints are safetensors
if checkpoint_files is not None and checkpoint_files[0].endswith(".safetensors"):
Expand Down