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

Fix: loading DBRX back from saved path #35728

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Fix: loading DBRX back from saved path #35728

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d72b9e9
fix dtype as dict for some models + add test
zucchini-nlp Jan 16, 2025
c3aecba
Merge branch 'main' into dtype-composite
zucchini-nlp Jan 16, 2025
b9effdb
Merge branch 'main' into dtype-composite
zucchini-nlp Jan 17, 2025
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
4 changes: 3 additions & 1 deletion src/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4029,7 +4029,9 @@ def from_pretrained(
sub_config = getattr(config, sub_config_key)
sub_config.torch_dtype = torch_dtype
elif isinstance(torch_dtype, torch.dtype):
pass
for sub_config_key in config.sub_configs.keys():
sub_config = getattr(config, sub_config_key)
sub_config.torch_dtype = torch_dtype
elif isinstance(torch_dtype, dict):
for key, curr_dtype in torch_dtype.items():
if hasattr(config, key):
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/dbrx/configuration_dbrx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,7 +57,7 @@ def __init__(
self.kv_n_heads = kv_n_heads
self.rope_theta = rope_theta

for k in ["model_type", "attn_implementation", "transformers_version", "_commit_hash"]:
for k in ["model_type", "attn_implementation", "transformers_version", "_commit_hash", "torch_dtype"]:
if k in kwargs:
kwargs.pop(k)
if len(kwargs) != 0:
Expand Down Expand Up @@ -109,7 +109,7 @@ def __init__(
self.moe_loss_weight = moe_loss_weight
self.moe_normalize_expert_weights = moe_normalize_expert_weights

for k in ["model_type", "attn_implementation", "transformers_version", "_commit_hash"]:
for k in ["model_type", "attn_implementation", "transformers_version", "_commit_hash", "torch_dtype"]:
if k in kwargs:
kwargs.pop(k)
if len(kwargs) != 0:
Expand Down
6 changes: 6 additions & 0 deletions tests/test_modeling_common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -331,6 +331,12 @@ def check_save_load(out1, out2):
with torch.no_grad():
second = model(**self._prepare_for_class(inputs_dict, model_class))[0]

# Save and load second time because `from_pretrained` adds a bunch of new config fields
# so we need to make sure those fields can be loaded back after saving
# Simply init as `model(config)` doesn't add those fields
model.save_pretrained(tmpdirname)
model = model_class.from_pretrained(tmpdirname)

if isinstance(first, tuple) and isinstance(second, tuple):
for tensor1, tensor2 in zip(first, second):
check_save_load(tensor1, tensor2)
Expand Down