[misc] lint code (#9395)

This commit is contained in:
Yaowei Zheng 2025-11-03 22:08:59 +08:00 committed by GitHub
parent 215580c77d
commit 3ae15da9c0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
17 changed files with 82 additions and 75 deletions

View File

@ -137,7 +137,6 @@ def _load_single_dataset(
cache_dir=model_args.cache_dir, cache_dir=model_args.cache_dir,
token=model_args.hf_hub_token, token=model_args.hf_hub_token,
num_proc=data_args.preprocessing_num_workers, num_proc=data_args.preprocessing_num_workers,
trust_remote_code=model_args.trust_remote_code,
streaming=data_args.streaming and dataset_attr.load_from != "file", streaming=data_args.streaming and dataset_attr.load_from != "file",
) )
if data_args.streaming and dataset_attr.load_from == "file": if data_args.streaming and dataset_attr.load_from == "file":

View File

@ -70,7 +70,6 @@ if TYPE_CHECKING:
from transformers.image_processing_utils import BaseImageProcessor from transformers.image_processing_utils import BaseImageProcessor
from transformers.video_processing_utils import BaseVideoProcessor from transformers.video_processing_utils import BaseVideoProcessor
class EncodedImage(TypedDict): class EncodedImage(TypedDict):
path: Optional[str] path: Optional[str]
bytes: Optional[bytes] bytes: Optional[bytes]

View File

@ -56,7 +56,18 @@ LAYERNORM_NAMES = {"norm", "ln"}
LLAMABOARD_CONFIG = "llamaboard_config.yaml" LLAMABOARD_CONFIG = "llamaboard_config.yaml"
MCA_SUPPORTED_MODELS = {"deepseek_v3", "llama", "mistral", "mixtral", "qwen2", "qwen2_vl", "qwen2_5_vl", "qwen3", "qwen3_moe", "qwen3_next"} MCA_SUPPORTED_MODELS = {
"deepseek_v3",
"llama",
"mistral",
"mixtral",
"qwen2",
"qwen2_vl",
"qwen2_5_vl",
"qwen3",
"qwen3_moe",
"qwen3_next",
}
METHODS = ["full", "freeze", "lora", "oft"] METHODS = ["full", "freeze", "lora", "oft"]

View File

@ -475,7 +475,12 @@ class FinetuningArguments(
) )
use_mca: bool = field( use_mca: bool = field(
default=False, default=False,
metadata={"help": "Whether or not to use MCA (Megatron Core Adapter) training. Controlled by USE_MCA environment variable."}, metadata={
"help": (
"Whether or not to use MCA (Megatron Core Adapter) training. "
"Controlled by USE_MCA environment variable."
)
},
) )
use_muon: bool = field( use_muon: bool = field(
default=False, default=False,

View File

@ -55,12 +55,16 @@ _EVAL_CLS = tuple[ModelArguments, DataArguments, EvaluationArguments, Finetuning
if is_mcore_adapter_available() and is_env_enabled("USE_MCA"): if is_mcore_adapter_available() and is_env_enabled("USE_MCA"):
from mcore_adapter import TrainingArguments as McaTrainingArguments from mcore_adapter import TrainingArguments as McaTrainingArguments
_TRAIN_MCA_ARGS = [ModelArguments, DataArguments, McaTrainingArguments, FinetuningArguments, GeneratingArguments] _TRAIN_MCA_ARGS = [ModelArguments, DataArguments, McaTrainingArguments, FinetuningArguments, GeneratingArguments]
_TRAIN_MCA_CLS = tuple[ModelArguments, DataArguments, McaTrainingArguments, FinetuningArguments, GeneratingArguments] _TRAIN_MCA_CLS = tuple[
ModelArguments, DataArguments, McaTrainingArguments, FinetuningArguments, GeneratingArguments
]
else: else:
_TRAIN_MCA_ARGS = [] _TRAIN_MCA_ARGS = []
_TRAIN_MCA_CLS = tuple() _TRAIN_MCA_CLS = tuple()
def read_args(args: Optional[Union[dict[str, Any], list[str]]] = None) -> Union[dict[str, Any], list[str]]: def read_args(args: Optional[Union[dict[str, Any], list[str]]] = None) -> Union[dict[str, Any], list[str]]:
r"""Get arguments from the command line or a config file.""" r"""Get arguments from the command line or a config file."""
if args is not None: if args is not None:

View File

@ -20,17 +20,18 @@ from transformers import Seq2SeqTrainingArguments
from transformers.training_args import _convert_str_dict from transformers.training_args import _convert_str_dict
from ..extras.misc import is_env_enabled, use_ray from ..extras.misc import is_env_enabled, use_ray
from ..extras.packages import is_mcore_adapter_available
if is_env_enabled("USE_MCA"): if is_env_enabled("USE_MCA"):
try: if not is_mcore_adapter_available():
from mcore_adapter import Seq2SeqTrainingArguments as McaSeq2SeqTrainingArguments
BaseTrainingArguments = McaSeq2SeqTrainingArguments
except ImportError:
raise ImportError( raise ImportError(
"mcore_adapter is required when USE_MCA=1.", "mcore_adapter is required when USE_MCA=1. Please install `mcore_adapter` and its dependencies."
"Please install `mcore_adapter` and its dependencies."
) )
from mcore_adapter import Seq2SeqTrainingArguments as McaSeq2SeqTrainingArguments
BaseTrainingArguments = McaSeq2SeqTrainingArguments
else: else:
BaseTrainingArguments = Seq2SeqTrainingArguments BaseTrainingArguments = Seq2SeqTrainingArguments

View File

@ -54,8 +54,7 @@ def launch():
) )
command = sys.argv.pop(1) if len(sys.argv) > 1 else "help" command = sys.argv.pop(1) if len(sys.argv) > 1 else "help"
if is_env_enabled("USE_MCA"): if is_env_enabled("USE_MCA"): # force use torchrun
# force use torchrun
os.environ["FORCE_TORCHRUN"] = "1" os.environ["FORCE_TORCHRUN"] = "1"
if command == "train" and (is_env_enabled("FORCE_TORCHRUN") or (get_device_count() > 1 and not use_ray())): if command == "train" and (is_env_enabled("FORCE_TORCHRUN") or (get_device_count() > 1 and not use_ray())):

View File

@ -16,4 +16,3 @@ from .workflow import run_dpo, run_pt, run_sft
__all__ = ["run_dpo", "run_pt", "run_sft"] __all__ = ["run_dpo", "run_pt", "run_sft"]

View File

@ -75,12 +75,17 @@ def _data_collator_wrapper(data_collator: Any):
return wrapper return wrapper
def _check_model_support(model_args: ModelArguments): def _check_model_support(model_args: ModelArguments):
from transformers import AutoConfig as HfAutoConfig from transformers import AutoConfig as HfAutoConfig
config = HfAutoConfig.from_pretrained(model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code)
config = HfAutoConfig.from_pretrained(
model_args.model_name_or_path, trust_remote_code=model_args.trust_remote_code
)
if config.model_type not in MCA_SUPPORTED_MODELS: if config.model_type not in MCA_SUPPORTED_MODELS:
raise ValueError(f"Model {config.model_type} is not supported by MCA.") raise ValueError(f"Model {config.model_type} is not supported by MCA.")
def run_pt( def run_pt(
model_args: ModelArguments, model_args: ModelArguments,
data_args: DataArguments, data_args: DataArguments,
@ -161,22 +166,23 @@ def run_sft(
model = AutoModel.from_pretrained(model_args.model_name_or_path, training_args) model = AutoModel.from_pretrained(model_args.model_name_or_path, training_args)
# optional freezing for qwen2_vl, qwen2_5_vl # optional freezing for qwen2_vl, qwen2_5_vl
if getattr(model.config, "hf_model_type", None) in ["qwen2_vl", "qwen2_5_vl"] and finetuning_args.freeze_vision_tower: if getattr(model.config, "hf_model_type", None) in ["qwen2_vl", "qwen2_5_vl"]:
for name, p in model.named_parameters(): params_to_freeze = []
if any(name.startswith(k) for k in ["vision_model.blocks", "vision_model.patch_embed"]): if finetuning_args.freeze_vision_tower:
p.requires_grad_(False) params_to_freeze.extend(["vision_model.blocks", "vision_model.patch_embed"])
if getattr(model.config, "hf_model_type", None) in ["qwen2_vl", "qwen2_5_vl"] and finetuning_args.freeze_multi_modal_projector:
for name, p in model.named_parameters():
if any(name.startswith(k) for k in ["multi_modal_projector"]):
p.requires_grad_(False)
if getattr(model.config, "hf_model_type", None) in ["qwen2_vl", "qwen2_5_vl"] and finetuning_args.freeze_language_model:
for name, p in model.named_parameters():
if any(name.startswith(k) for k in ["embedding", "decoder", "output_layer"]):
p.requires_grad_(False)
pad_to_max = ( if finetuning_args.freeze_multi_modal_projector:
training_args.expert_model_parallel_size is not None and training_args.expert_model_parallel_size > 1 params_to_freeze.extend(["multi_modal_projector"])
)
if finetuning_args.freeze_language_model:
params_to_freeze.extend(["embedding", "decoder", "output_layer"])
if params_to_freeze:
for name, p in model.named_parameters():
if any(name.startswith(k) for k in params_to_freeze):
p.requires_grad_(False)
pad_to_max = training_args.expert_model_parallel_size is not None and training_args.expert_model_parallel_size > 1
data_collator = SFTDataCollatorWith4DAttentionMask( data_collator = SFTDataCollatorWith4DAttentionMask(
template=template, template=template,
padding="max_length" if pad_to_max else "longest", padding="max_length" if pad_to_max else "longest",
@ -239,9 +245,7 @@ def run_dpo(
dataset_module = get_dataset(template, model_args, data_args, training_args, stage="rm", **tokenizer_module) dataset_module = get_dataset(template, model_args, data_args, training_args, stage="rm", **tokenizer_module)
data_args.cutoff_len -= 1 data_args.cutoff_len -= 1
pad_to_max = ( pad_to_max = training_args.expert_model_parallel_size is not None and training_args.expert_model_parallel_size > 1
training_args.expert_model_parallel_size is not None and training_args.expert_model_parallel_size > 1
)
dpo_config = DPOConfig( dpo_config = DPOConfig(
beta=finetuning_args.pref_beta, beta=finetuning_args.pref_beta,
pref_loss=finetuning_args.pref_loss, pref_loss=finetuning_args.pref_loss,
@ -289,4 +293,3 @@ def run_dpo(
keys += ["eval_loss"] keys += ["eval_loss"]
plot_loss(training_args.output_dir, keys=keys) plot_loss(training_args.output_dir, keys=keys)

View File

@ -71,13 +71,17 @@ def _training_function(config: dict[str, Any]) -> None:
raise ImportError("mcore_adapter is not installed. Please install it with `pip install mcore-adapter`.") raise ImportError("mcore_adapter is not installed. Please install it with `pip install mcore-adapter`.")
if finetuning_args.stage == "pt": if finetuning_args.stage == "pt":
from .mca import run_pt as run_pt_mca from .mca import run_pt as run_pt_mca
run_pt_mca(model_args, data_args, training_args, finetuning_args, callbacks) run_pt_mca(model_args, data_args, training_args, finetuning_args, callbacks)
elif finetuning_args.stage == "sft": elif finetuning_args.stage == "sft":
from .mca import run_sft as run_sft_mca from .mca import run_sft as run_sft_mca
run_sft_mca(model_args, data_args, training_args, finetuning_args, callbacks) run_sft_mca(model_args, data_args, training_args, finetuning_args, callbacks)
else: # dpo elif finetuning_args.stage == "dpo":
from .mca import run_dpo as run_dpo_mca from .mca import run_dpo as run_dpo_mca
run_dpo_mca(model_args, data_args, training_args, finetuning_args, callbacks) run_dpo_mca(model_args, data_args, training_args, finetuning_args, callbacks)
elif finetuning_args.stage == "pt": elif finetuning_args.stage == "pt":
run_pt(model_args, data_args, training_args, finetuning_args, callbacks) run_pt(model_args, data_args, training_args, finetuning_args, callbacks)
elif finetuning_args.stage == "sft": elif finetuning_args.stage == "sft":

View File

@ -24,7 +24,7 @@ class KernelType(str, Enum):
class DeviceType(str, Enum): class DeviceType(str, Enum):
CPU = 'cpu' CPU = "cpu"
CUDA = 'cuda' CUDA = "cuda"
NPU = 'npu' NPU = "npu"
XPU = 'xpu' XPU = "xpu"

View File

@ -27,14 +27,11 @@ def _npu_swiglu_forward(self, hidden_state):
import torch_npu import torch_npu
return self.down_proj( return self.down_proj(
torch_npu.npu_swiglu( torch_npu.npu_swiglu(torch.cat((self.gate_proj(hidden_state), self.up_proj(hidden_state)), dim=-1), dim=-1)
torch.cat((self.gate_proj(hidden_state), self.up_proj(hidden_state)), dim=-1), dim=-1
)
) )
class NpuSwiGluKernel(MetaSwiGluKernel): class NpuSwiGluKernel(MetaSwiGluKernel):
device = DeviceType.NPU device = DeviceType.NPU
kernel = _npu_swiglu_forward kernel = _npu_swiglu_forward
@ -43,7 +40,7 @@ class NpuSwiGluKernel(MetaSwiGluKernel):
KERNEL_REGISTRY.register(kernel_type, device_type, cls) KERNEL_REGISTRY.register(kernel_type, device_type, cls)
@classmethod @classmethod
def apply(cls, model, **kwargs) -> 'HFModel': def apply(cls, model, **kwargs) -> "HFModel":
if not is_torch_npu_available(): if not is_torch_npu_available():
return model return model
@ -51,7 +48,6 @@ class NpuSwiGluKernel(MetaSwiGluKernel):
for name, module in model.named_modules(): for name, module in model.named_modules():
# Match any module whose class name contains "RMSNorm" # Match any module whose class name contains "RMSNorm"
if re.search(swiglu_pattern, module.__class__.__name__): if re.search(swiglu_pattern, module.__class__.__name__):
# Bind function as an instance method to preserve `self` semantics # Bind function as an instance method to preserve `self` semantics
# and replace the original forward # and replace the original forward
module.forward = types.MethodType(cls.kernel, module) module.forward = types.MethodType(cls.kernel, module)

View File

@ -21,10 +21,10 @@ from .constants import DeviceType, KernelType
class KernelRegistry: class KernelRegistry:
_instance: Optional['KernelRegistry'] = None _instance: Optional["KernelRegistry"] = None
_initialized: bool = False _initialized: bool = False
def __new__(cls, *args: Any, **kwargs: Any) -> 'KernelRegistry': def __new__(cls, *args: Any, **kwargs: Any) -> "KernelRegistry":
if cls._instance is None: if cls._instance is None:
cls._instance = super().__new__(cls) cls._instance = super().__new__(cls)
return cls._instance return cls._instance
@ -36,10 +36,7 @@ class KernelRegistry:
self._initialized = True self._initialized = True
def register( def register(
self, self, kernel_type: KernelType, device_type: DeviceType, kernel_impl: Optional[Callable[..., Any]]
kernel_type: KernelType,
device_type: DeviceType,
kernel_impl: Optional[Callable[..., Any]]
) -> None: ) -> None:
"""Register a kernel implementation. """Register a kernel implementation.
@ -57,11 +54,7 @@ class KernelRegistry:
self._registry[kernel_type][device_type] = kernel_impl self._registry[kernel_type][device_type] = kernel_impl
print(f"Registered kernel {kernel_type.name} for device {device_type.name}.") print(f"Registered kernel {kernel_type.name} for device {device_type.name}.")
def get_kernel( def get_kernel(self, kernel_type: KernelType, device_type: DeviceType) -> Optional[Callable[..., Any]]:
self,
kernel_type: KernelType,
device_type: DeviceType
) -> Optional[Callable[..., Any]]:
return self._registry.get(kernel_type, {}).get(device_type) return self._registry.get(kernel_type, {}).get(device_type)
@ -84,35 +77,30 @@ class MetaKernel(ABC):
class MetaFlashAttentionKernel(MetaKernel): class MetaFlashAttentionKernel(MetaKernel):
@classmethod @classmethod
def apply(cls, model: HFModel, **kwargs) -> HFModel: def apply(cls, model: HFModel, **kwargs) -> HFModel:
raise NotImplementedError raise NotImplementedError
class MetaRMSNormKernel(MetaKernel): class MetaRMSNormKernel(MetaKernel):
@classmethod @classmethod
def apply(cls, model: HFModel, **kwargs) -> HFModel: def apply(cls, model: HFModel, **kwargs) -> HFModel:
raise NotImplementedError raise NotImplementedError
class MetaSwiGluKernel(MetaKernel): class MetaSwiGluKernel(MetaKernel):
@classmethod @classmethod
def apply(cls, model: HFModel, **kwargs) -> HFModel: def apply(cls, model: HFModel, **kwargs) -> HFModel:
raise NotImplementedError raise NotImplementedError
class MetaRoPEKernel(MetaKernel): class MetaRoPEKernel(MetaKernel):
@classmethod @classmethod
def apply(cls, model: HFModel, **kwargs) -> HFModel: def apply(cls, model: HFModel, **kwargs) -> HFModel:
raise NotImplementedError raise NotImplementedError
class MetaMoEKernel(MetaKernel): class MetaMoEKernel(MetaKernel):
@classmethod @classmethod
def apply(cls, model: HFModel, **kwargs) -> HFModel: def apply(cls, model: HFModel, **kwargs) -> HFModel:
raise NotImplementedError raise NotImplementedError
@ -130,7 +118,7 @@ def discover_kernels(model: HFModel) -> list[MetaKernel]:
return [] return []
def apply_kernel(model: HFModel, kernel: type[MetaKernel], /, **kwargs) -> 'HFModel': def apply_kernel(model: HFModel, kernel: type[MetaKernel], /, **kwargs) -> "HFModel":
"""Call the MetaKernel's `apply` to perform the replacement. """Call the MetaKernel's `apply` to perform the replacement.
Corresponding replacement logic is maintained inside each kernel; the only Corresponding replacement logic is maintained inside each kernel; the only
@ -145,4 +133,6 @@ def apply_kernel(model: HFModel, kernel: type[MetaKernel], /, **kwargs) -> 'HFMo
if issubclass(kernel, MetaKernel) and kernel.device == get_available_accelerator().type: if issubclass(kernel, MetaKernel) and kernel.device == get_available_accelerator().type:
return kernel.apply(model, **kwargs) return kernel.apply(model, **kwargs)
raise ValueError(f"{kernel} must be a MetaKernel instance, or the kernel don't match the device type. got {kernel.device} and {get_available_accelerator().type} instead.") raise ValueError(
f"{kernel} must be a MetaKernel instance, or the kernel don't match the device type. got {kernel.device} and {get_available_accelerator().type} instead."
)

View File

@ -65,7 +65,6 @@ class NpuRMSNormKernel(MetaRMSNormKernel):
for name, module in model.named_modules(): for name, module in model.named_modules():
# Match any module whose class name contains "RMSNorm" # Match any module whose class name contains "RMSNorm"
if re.search(rms_norm_pattern, module.__class__.__name__): if re.search(rms_norm_pattern, module.__class__.__name__):
# Bind function as an instance method to preserve `self` semantics # Bind function as an instance method to preserve `self` semantics
# and replace the original forward # and replace the original forward
module.forward = types.MethodType(cls.kernel, module) module.forward = types.MethodType(cls.kernel, module)

View File

@ -59,7 +59,7 @@ class NpuRoPEKernel(MetaRoPEKernel):
KERNEL_REGISTRY.register(kernel_type, device_type, cls) KERNEL_REGISTRY.register(kernel_type, device_type, cls)
@classmethod @classmethod
def apply(cls, model, **kwargs) -> 'HFModel': def apply(cls, model, **kwargs) -> "HFModel":
"""Apply RoPE acceleration by monkey-patching `apply_rotary_pos_emb`. """Apply RoPE acceleration by monkey-patching `apply_rotary_pos_emb`.
This function iterates through the model's modules to find attention layers, This function iterates through the model's modules to find attention layers,
@ -96,7 +96,7 @@ class NpuQwen2VLRoPEKernel(MetaRoPEKernel):
KERNEL_REGISTRY.register(kernel_type, device_type, cls) KERNEL_REGISTRY.register(kernel_type, device_type, cls)
@classmethod @classmethod
def apply(cls, model, **kwargs) -> 'HFModel': def apply(cls, model, **kwargs) -> "HFModel":
"""Apply RoPE acceleration by monkey-patching `apply_rotary_pos_emb`. """Apply RoPE acceleration by monkey-patching `apply_rotary_pos_emb`.
This function iterates through the model's modules to find attention layers, This function iterates through the model's modules to find attention layers,

View File

@ -23,25 +23,25 @@ def get_available_accelerator():
""" """
accelerator = torch.accelerator.current_accelerator() accelerator = torch.accelerator.current_accelerator()
if accelerator is None: if accelerator is None:
return torch.device('cpu') return torch.device("cpu")
return accelerator return accelerator
@lru_cache @lru_cache
def is_torch_npu_available(): def is_torch_npu_available():
return get_available_accelerator().type == 'npu' return get_available_accelerator().type == "npu"
@lru_cache @lru_cache
def is_torch_cuda_available(): def is_torch_cuda_available():
return get_available_accelerator().type == 'cuda' return get_available_accelerator().type == "cuda"
@lru_cache @lru_cache
def is_torch_xpu_available(): def is_torch_xpu_available():
return get_available_accelerator().type == 'xpu' return get_available_accelerator().type == "xpu"
@lru_cache @lru_cache
def is_torch_mps_available(): def is_torch_mps_available():
return get_available_accelerator().type == 'mps' return get_available_accelerator().type == "mps"

View File

@ -19,11 +19,10 @@ from transformers import AutoModelForCausalLM
class TestKernelPlugin(unittest.TestCase): class TestKernelPlugin(unittest.TestCase):
@patch("torch.accelerator.current_accelerator")
@patch('torch.accelerator.current_accelerator')
def test_apply_kernel(self, mock_get_accelerator): def test_apply_kernel(self, mock_get_accelerator):
mock_device = MagicMock() mock_device = MagicMock()
mock_device.type = 'npu' mock_device.type = "npu"
mock_get_accelerator.return_value = mock_device mock_get_accelerator.return_value = mock_device
model = AutoModelForCausalLM.from_pretrained("llamafactory/tiny-random-qwen2.5") model = AutoModelForCausalLM.from_pretrained("llamafactory/tiny-random-qwen2.5")
@ -31,7 +30,6 @@ class TestKernelPlugin(unittest.TestCase):
original_rmsnorm_forward = model.model.layers[0].input_layernorm.forward original_rmsnorm_forward = model.model.layers[0].input_layernorm.forward
original_swiglu_forward = model.model.layers[0].mlp.forward original_swiglu_forward = model.model.layers[0].mlp.forward
from llamafactory.v1.plugins.model_plugins.kernels.mlp import npu_swiglu from llamafactory.v1.plugins.model_plugins.kernels.mlp import npu_swiglu
from llamafactory.v1.plugins.model_plugins.kernels.registry import apply_kernel from llamafactory.v1.plugins.model_plugins.kernels.registry import apply_kernel
from llamafactory.v1.plugins.model_plugins.kernels.rms_norm import npu_rms_norm from llamafactory.v1.plugins.model_plugins.kernels.rms_norm import npu_rms_norm