SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

fix: Removed duplicate field definitions in some classes (#31888)

Browse Source 
Removed duplicate field definitions in classes.
This commit is contained in:
Sai-Suraj-27
2024-07-10 18:16:31 +05:30
committed by GitHub
parent 9d98706b3f
commit da79b18087
8 changed files with 0 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
examples/flax/language-modeling/run_clm_flax.py
View File

@@ -225,9 +225,6 @@ class DataTrainingArguments:
) )
}, },
) )
overwrite_cache: bool = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
)
validation_split_percentage: Optional[int] = field( validation_split_percentage: Optional[int] = field(
default=5, default=5,
metadata={ metadata={

3
examples/research_projects/jax-projects/hybrid_clip/run_hybrid_clip.py
View File

@@ -163,9 +163,6 @@ class DataTrainingArguments:
overwrite_cache: bool = field( overwrite_cache: bool = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets"} default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
) )
overwrite_cache: bool = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
)
preprocessing_num_workers: Optional[int] = field( preprocessing_num_workers: Optional[int] = field(
default=None, default=None,
metadata={"help": "The number of processes to use for the preprocessing."}, metadata={"help": "The number of processes to use for the preprocessing."},

3
examples/research_projects/jax-projects/model_parallel/run_clm_mp.py
View File

@@ -156,9 +156,6 @@ class DataTrainingArguments:
) )
}, },
) )
overwrite_cache: bool = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
)
preprocessing_num_workers: Optional[int] = field( preprocessing_num_workers: Optional[int] = field(
default=None, default=None,
metadata={"help": "The number of processes to use for the preprocessing."}, metadata={"help": "The number of processes to use for the preprocessing."},

1
src/transformers/models/deformable_detr/modeling_deformable_detr.py
View File

@@ -1080,7 +1080,6 @@ class DeformableDetrPreTrainedModel(PreTrainedModel):
main_input_name = "pixel_values" main_input_name = "pixel_values"
supports_gradient_checkpointing = True supports_gradient_checkpointing = True
_no_split_modules = [r"DeformableDetrConvEncoder", r"DeformableDetrEncoderLayer", r"DeformableDetrDecoderLayer"] _no_split_modules = [r"DeformableDetrConvEncoder", r"DeformableDetrEncoderLayer", r"DeformableDetrDecoderLayer"]
supports_gradient_checkpointing = True
def _init_weights(self, module): def _init_weights(self, module):
std = self.config.init_std std = self.config.init_std

1
src/transformers/models/video_llava/modeling_video_llava.py
View File

@@ -126,7 +126,6 @@ class VideoLlavaPreTrainedModel(PreTrainedModel):
_no_split_modules = ["VideoLlavaVisionAttention"] _no_split_modules = ["VideoLlavaVisionAttention"]
_skip_keys_device_placement = "past_key_values" _skip_keys_device_placement = "past_key_values"
_supports_flash_attn_2 = True _supports_flash_attn_2 = True
_no_split_modules = ["VideoLlavaVisionAttention"]
def _init_weights(self, module): def _init_weights(self, module):
std = ( std = (

1
tests/models/fnet/test_modeling_fnet.py
View File

@@ -295,7 +295,6 @@ class FNetModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
# Skip Tests # Skip Tests
test_pruning = False test_pruning = False
test_head_masking = False test_head_masking = False
test_pruning = False
# TODO: Fix the failed tests # TODO: Fix the failed tests
def is_pipeline_test_to_skip( def is_pipeline_test_to_skip(

1
tests/models/mamba/test_modeling_mamba.py
View File

@@ -258,7 +258,6 @@ class MambaModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixi
test_model_parallel = False test_model_parallel = False
test_pruning = False test_pruning = False
test_head_masking = False # Mamba does not have attention heads test_head_masking = False # Mamba does not have attention heads
test_model_parallel = False
pipeline_model_mapping = ( pipeline_model_mapping = (
{"feature-extraction": MambaModel, "text-generation": MambaForCausalLM} if is_torch_available() else {} {"feature-extraction": MambaModel, "text-generation": MambaForCausalLM} if is_torch_available() else {}
) )

1
tests/models/recurrent_gemma/test_modeling_recurrent_gemma.py
View File

@@ -298,7 +298,6 @@ class RecurrentGemmaModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineT
test_model_parallel = False test_model_parallel = False
test_pruning = False test_pruning = False
test_head_masking = False # RecurrentGemma does not have attention heads test_head_masking = False # RecurrentGemma does not have attention heads
test_model_parallel = False
# Need to remove 0.9 in `test_cpu_offload` # Need to remove 0.9 in `test_cpu_offload`
# This is because we are hitting edge cases with the causal_mask buffer # This is because we are hitting edge cases with the causal_mask buffer