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

update codegen config for support codegen2 #720

Merged
merged 9 commits into from
May 21, 2024
Merged

update codegen config for support codegen2 #720

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
a85eae6
update codegen config for support codegen2
eaidova May 20, 2024
6e3b010
add support DBRX
eaidova May 20, 2024
d1782d0
add qwen2moe support
eaidova May 20, 2024
37f2094
fix test models
eaidova May 21, 2024
a255a08
buichuan sdpa
eaidova May 21, 2024
b64a5c8
apply review comments
eaidova May 21, 2024
0ecf082
Apply suggestions from code review
eaidova May 21, 2024
9984045
Apply suggestions from code review
eaidova May 21, 2024
0b380cf
resolve conflict
eaidova May 21, 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
47 changes: 47 additions & 0 deletions optimum/exporters/openvino/model_configs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@

from optimum.exporters.onnx.config import TextDecoderOnnxConfig, TextDecoderWithPositionIdsOnnxConfig
from optimum.exporters.onnx.model_configs import (
CodeGenOnnxConfig,
FalconOnnxConfig,
GemmaOnnxConfig,
LlamaOnnxConfig,
Expand All @@ -44,6 +45,8 @@
AquilaModelPatcher,
BaichuanModelPatcher,
ChatGLMModelPatcher,
CodeGenModelPatcher,
DBRXModelPatcher,
GemmaModelPatcher,
InternLM2Patcher,
InternLMModelPatcher,
Expand Down Expand Up @@ -112,6 +115,15 @@ class Qwen2OpenVINOConfig(TextDecoderWithPositionIdsOnnxConfig):
NORMALIZED_CONFIG_CLASS = NormalizedTextConfig


@register_in_tasks_manager("qwen2-moe", *["text-generation", "text-generation-with-past"], library_name="transformers")
class Qwen2MoEOpenVINOConfig(TextDecoderWithPositionIdsOnnxConfig):
DEFAULT_ONNX_OPSET = 14

DUMMY_INPUT_GENERATOR_CLASSES = (DummyTextInputGenerator, MistralDummyPastKeyValuesGenerator)
DUMMY_PKV_GENERATOR_CLASS = MistralDummyPastKeyValuesGenerator
NORMALIZED_CONFIG_CLASS = NormalizedTextConfig


@register_in_tasks_manager("minicpm", *["text-generation", "text-generation-with-past"], library_name="transformers")
class MiniCPMOpenVINOConfig(TextDecoderWithPositionIdsOnnxConfig):
DEFAULT_ONNX_OPSET = 14
Expand Down Expand Up @@ -738,3 +750,38 @@ def patch_model_for_export(
self, model: Union["PreTrainedModel", "TFPreTrainedModel"], model_kwargs: Optional[Dict[str, Any]] = None
) -> "ModelPatcher":
return InternLMModelPatcher(self, model, model_kwargs=model_kwargs)


@register_in_tasks_manager(
"codegen",
*["feature-extraction", "feature-extraction-with-past", "text-generation", "text-generation-with-past"],
library_name="transformers",
)
class CodeGenOpenVINOConfig(CodeGenOnnxConfig):
def patch_model_for_export(
self, model: Union["PreTrainedModel", "TFPreTrainedModel"], model_kwargs: Optional[Dict[str, Any]] = None
) -> "ModelPatcher":
return CodeGenModelPatcher(self, model, model_kwargs=model_kwargs)


@register_in_tasks_manager(
"dbrx",
*["text-generation", "text-generation-with-past"],
library_name="transformers",
)
class DBRXOpenVINOConfig(TextDecoderWithPositionIdsOnnxConfig):
DEFAULT_ONNX_OPSET = 14
NORMALIZED_CONFIG_CLASS = NormalizedTextConfig.with_args(
num_attention_heads="n_heads",
hidden_size="d_model",
num_layers="n_layers",
num_key_value_heads="attn_config.kv_n_heads",
allow_new=True,
)
DUMMY_INPUT_GENERATOR_CLASSES = (DummyTextInputGenerator, MistralDummyPastKeyValuesGenerator)
DUMMY_PKV_GENERATOR_CLASS = MistralDummyPastKeyValuesGenerator

def patch_model_for_export(
self, model: Union["PreTrainedModel", "TFPreTrainedModel"], model_kwargs: Optional[Dict[str, Any]] = None
) -> "ModelPatcher":
return DBRXModelPatcher(self, model, model_kwargs=model_kwargs)
Loading
Loading