Skip to content

Fix test_llama #11165

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
wants to merge 1 commit into from
Closed

Fix test_llama #11165

Show file tree
Hide file tree
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
4 changes: 3 additions & 1 deletion backends/arm/test/models/test_llama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
TosaPipelineMI,
)

from executorch.examples.models.llama.config.llm_config_utils import convert_args_to_llm_config
from executorch.examples.models.llama.export_llama_lib import (
build_args_parser,
get_llama_model,
Expand Down Expand Up @@ -89,8 +90,9 @@ def prepare_model(self):
]
parser = build_args_parser()
args = parser.parse_args(args)
llm_config = convert_args_to_llm_config(args)

llama_model, llama_inputs, llama_meta = get_llama_model(args)
llama_model, llama_inputs, llama_meta = get_llama_model(llm_config)

return llama_model, llama_inputs, llama_meta

Expand Down
10 changes: 3 additions & 7 deletions examples/models/llama/export_llama_lib.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -805,10 +805,6 @@ def _qmode_type(value):


def _validate_args(llm_config):
"""
TODO: Combine all the backends under --backend args
"""

if llm_config.export.max_context_length < llm_config.export.max_seq_length:
raise ValueError(
f"max_context_length {llm_config.export.max_context_length} must be >= max_seq_len {llm_config.export.max_seq_length}. max_context_length impacts kv cache size that is used to remember history, while max_seq_length refers to user prompt length. Please use --max_context_length to specify context length."
Expand Down Expand Up @@ -1498,9 +1494,9 @@ def _get_source_transforms( # noqa
return transforms


def get_llama_model(args):
_validate_args(args)
e_mgr = _prepare_for_llama_export(args)
def get_llama_model(llm_config: LlmConfig):
_validate_args(llm_config)
e_mgr = _prepare_for_llama_export(llm_config)
model = (
e_mgr.model.eval().to(device="cuda")
if torch.cuda.is_available()
Expand Down
Loading