Skip to content

GitLab

"benchmark/git@developer.sourcefind.cn:change/sglang.git" did not exist on "9a71500cfb267bef4f0c2bfc4ce60eba9fe5674d"
Unverified Commit c6f7ea19 authored by Mowaninuola Osifeso's avatar Mowaninuola Osifeso Committed by GitHub
Browse files

Add type hints to xlnet (#16214) 

* added type hints to xlnet PT

* added type hints to xlnet TF

* added type hints to xlnet TF
parent abf3cc70
Hide whitespace changes
Inline Side-by-side
Showing with 194 additions and 192 deletions
src/transformers/models/xlnet/modeling_tf_xlnet.py
View file @ c6f7ea19
...@@ -19,8 +19,9 @@ ...@@ -19,8 +19,9 @@
import warnings import warnings
from dataclasses import dataclass from dataclasses import dataclass
from typing import List, Optional, Tuple from typing import List, Optional, Tuple, Union
import numpy as np
import tensorflow as tf import tensorflow as tf
from ...activations_tf import get_tf_activation from ...activations_tf import get_tf_activation
...@@ -34,6 +35,7 @@ from ...file_utils import ( ...@@ -34,6 +35,7 @@ from ...file_utils import (
) )
from ...modeling_tf_utils import ( from ...modeling_tf_utils import (
TFCausalLanguageModelingLoss, TFCausalLanguageModelingLoss,
TFModelInputType,
TFMultipleChoiceLoss, TFMultipleChoiceLoss,
TFPreTrainedModel, TFPreTrainedModel,
TFQuestionAnsweringLoss, TFQuestionAnsweringLoss,
...@@ -1245,23 +1247,23 @@ class TFXLNetLMHeadModel(TFXLNetPreTrainedModel, TFCausalLanguageModelingLoss): ...@@ -1245,23 +1247,23 @@ class TFXLNetLMHeadModel(TFXLNetPreTrainedModel, TFCausalLanguageModelingLoss):
@replace_return_docstrings(output_type=TFXLNetLMHeadModelOutput, config_class=_CONFIG_FOR_DOC) @replace_return_docstrings(output_type=TFXLNetLMHeadModelOutput, config_class=_CONFIG_FOR_DOC)
def call( def call(
self, self,
input_ids=None, input_ids: Optional[TFModelInputType] = None,
attention_mask=None, attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
mems=None, mems: Optional[Union[np.ndarray, tf.Tensor]] = None,
perm_mask=None, perm_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
target_mapping=None, target_mapping: Optional[Union[np.ndarray, tf.Tensor]] = None,
token_type_ids=None, token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
input_mask=None, input_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
head_mask=None, head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
inputs_embeds=None, inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
labels=None, labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
training=False, training: Optional[bool] = False,
**kwargs, **kwargs,
): ) -> Union[TFXLNetLMHeadModelOutput, Tuple[tf.Tensor]]:
r""" r"""
labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*): labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
Labels for computing the cross entropy classification loss. Indices should be in `[0, ..., Labels for computing the cross entropy classification loss. Indices should be in `[0, ...,
...@@ -1377,23 +1379,23 @@ class TFXLNetForSequenceClassification(TFXLNetPreTrainedModel, TFSequenceClassif ...@@ -1377,23 +1379,23 @@ class TFXLNetForSequenceClassification(TFXLNetPreTrainedModel, TFSequenceClassif
) )
def call( def call(
self, self,
input_ids=None, input_ids: Optional[TFModelInputType] = None,
attention_mask=None, attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
mems=None, mems: Optional[Union[np.ndarray, tf.Tensor]] = None,
perm_mask=None, perm_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
target_mapping=None, target_mapping: Optional[Union[np.ndarray, tf.Tensor]] = None,
token_type_ids=None, token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
input_mask=None, input_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
head_mask=None, head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
inputs_embeds=None, inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
labels=None, labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
training=False, training: Optional[bool] = False,
**kwargs, **kwargs,
): ) -> Union[TFXLNetForSequenceClassificationOutput, Tuple[tf.Tensor]]:
r""" r"""
labels (`tf.Tensor` of shape `(batch_size,)`, *optional*): labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
...@@ -1484,23 +1486,23 @@ class TFXLNetForMultipleChoice(TFXLNetPreTrainedModel, TFMultipleChoiceLoss): ...@@ -1484,23 +1486,23 @@ class TFXLNetForMultipleChoice(TFXLNetPreTrainedModel, TFMultipleChoiceLoss):
) )
def call( def call(
self, self,
input_ids=None, input_ids: Optional[TFModelInputType] = None,
token_type_ids=None, token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
input_mask=None, input_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
attention_mask=None, attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
mems=None, mems: Optional[Union[np.ndarray, tf.Tensor]] = None,
perm_mask=None, perm_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
target_mapping=None, target_mapping: Optional[Union[np.ndarray, tf.Tensor]] = None,
head_mask=None, head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
inputs_embeds=None, inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
labels=None, labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
training=False, training: Optional[bool] = False,
**kwargs, **kwargs,
): ) -> Union[TFXLNetForMultipleChoiceOutput, Tuple[tf.Tensor]]:
r""" r"""
labels (`tf.Tensor` of shape `(batch_size,)`, *optional*): labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]` Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]`
...@@ -1606,23 +1608,23 @@ class TFXLNetForTokenClassification(TFXLNetPreTrainedModel, TFTokenClassificatio ...@@ -1606,23 +1608,23 @@ class TFXLNetForTokenClassification(TFXLNetPreTrainedModel, TFTokenClassificatio
) )
def call( def call(
self, self,
input_ids=None, input_ids: Optional[TFModelInputType] = None,
attention_mask=None, attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
mems=None, mems: Optional[Union[np.ndarray, tf.Tensor]] = None,
perm_mask=None, perm_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
target_mapping=None, target_mapping: Optional[Union[np.ndarray, tf.Tensor]] = None,
token_type_ids=None, token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
input_mask=None, input_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
head_mask=None, head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
inputs_embeds=None, inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
labels=None, labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
training=False, training: Optional[bool] = False,
**kwargs, **kwargs,
): ) -> Union[TFXLNetForTokenClassificationOutput, Tuple[tf.Tensor]]:
r""" r"""
labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*): labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`. Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
...@@ -1693,24 +1695,24 @@ class TFXLNetForQuestionAnsweringSimple(TFXLNetPreTrainedModel, TFQuestionAnswer ...@@ -1693,24 +1695,24 @@ class TFXLNetForQuestionAnsweringSimple(TFXLNetPreTrainedModel, TFQuestionAnswer
) )
def call( def call(
self, self,
input_ids=None, input_ids: Optional[TFModelInputType] = None,
attention_mask=None, attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
mems=None, mems: Optional[Union[np.ndarray, tf.Tensor]] = None,
perm_mask=None, perm_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
target_mapping=None, target_mapping: Optional[Union[np.ndarray, tf.Tensor]] = None,
token_type_ids=None, token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
input_mask=None, input_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
head_mask=None, head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
inputs_embeds=None, inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
start_positions=None, start_positions: Optional[Union[np.ndarray, tf.Tensor]] = None,
end_positions=None, end_positions: Optional[Union[np.ndarray, tf.Tensor]] = None,
training=False, training: Optional[bool] = False,
**kwargs, **kwargs,
): ) -> Union[TFXLNetForQuestionAnsweringSimpleOutput, Tuple[tf.Tensor]]:
r""" r"""
start_positions (`tf.Tensor` of shape `(batch_size,)`, *optional*): start_positions (`tf.Tensor` of shape `(batch_size,)`, *optional*):
Labels for position (index) of the start of the labelled span for computing the token classification loss. Labels for position (index) of the start of the labelled span for computing the token classification loss.
......
src/transformers/models/xlnet/modeling_xlnet.py
View file @ c6f7ea19
...@@ -18,7 +18,7 @@ ...@@ -18,7 +18,7 @@
""" """
import warnings import warnings
from dataclasses import dataclass from dataclasses import dataclass
from typing import List, Optional, Tuple from typing import List, Optional, Tuple, Union
import torch import torch
from torch import nn from torch import nn
...@@ -1074,21 +1074,21 @@ class XLNetModel(XLNetPreTrainedModel): ...@@ -1074,21 +1074,21 @@ class XLNetModel(XLNetPreTrainedModel):
) )
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete after depreciation warning is removed **kwargs, # delete after depreciation warning is removed
): ) -> Union[Tuple, XLNetModelOutput]:
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
output_hidden_states = ( output_hidden_states = (
...@@ -1364,22 +1364,22 @@ class XLNetLMHeadModel(XLNetPreTrainedModel): ...@@ -1364,22 +1364,22 @@ class XLNetLMHeadModel(XLNetPreTrainedModel):
@replace_return_docstrings(output_type=XLNetLMHeadModelOutput, config_class=_CONFIG_FOR_DOC) @replace_return_docstrings(output_type=XLNetLMHeadModelOutput, config_class=_CONFIG_FOR_DOC)
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
labels=None, labels: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetLMHeadModelOutput]:
r""" r"""
labels (`torch.LongTensor` of shape `(batch_size, num_predict)`, *optional*): labels (`torch.LongTensor` of shape `(batch_size, num_predict)`, *optional*):
Labels for masked language modeling. `num_predict` corresponds to `target_mapping.shape[1]`. If Labels for masked language modeling. `num_predict` corresponds to `target_mapping.shape[1]`. If
...@@ -1526,22 +1526,22 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel): ...@@ -1526,22 +1526,22 @@ class XLNetForSequenceClassification(XLNetPreTrainedModel):
) )
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
labels=None, labels: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetForSequenceClassificationOutput]:
r""" r"""
labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
Labels for computing the sequence classification/regression loss. Indices should be in `[0, ..., Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
...@@ -1634,22 +1634,22 @@ class XLNetForTokenClassification(XLNetPreTrainedModel): ...@@ -1634,22 +1634,22 @@ class XLNetForTokenClassification(XLNetPreTrainedModel):
) )
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
labels=None, labels: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetForTokenClassificationOutput]:
r""" r"""
labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]` Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]`
...@@ -1722,22 +1722,22 @@ class XLNetForMultipleChoice(XLNetPreTrainedModel): ...@@ -1722,22 +1722,22 @@ class XLNetForMultipleChoice(XLNetPreTrainedModel):
) )
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
labels=None, labels: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetForMultipleChoiceOutput]:
r""" r"""
labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
...@@ -1826,23 +1826,23 @@ class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel): ...@@ -1826,23 +1826,23 @@ class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel):
) )
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
start_positions=None, start_positions: Optional[torch.Tensor] = None,
end_positions=None, end_positions: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetForQuestionAnsweringSimpleOutput]:
r""" r"""
start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
Labels for position (index) of the start of the labelled span for computing the token classification loss. Labels for position (index) of the start of the labelled span for computing the token classification loss.
...@@ -1935,26 +1935,26 @@ class XLNetForQuestionAnswering(XLNetPreTrainedModel): ...@@ -1935,26 +1935,26 @@ class XLNetForQuestionAnswering(XLNetPreTrainedModel):
@replace_return_docstrings(output_type=XLNetForQuestionAnsweringOutput, config_class=_CONFIG_FOR_DOC) @replace_return_docstrings(output_type=XLNetForQuestionAnsweringOutput, config_class=_CONFIG_FOR_DOC)
def forward( def forward(
self, self,
input_ids=None, input_ids: Optional[torch.Tensor] = None,
attention_mask=None, attention_mask: Optional[torch.Tensor] = None,
mems=None, mems: Optional[torch.Tensor] = None,
perm_mask=None, perm_mask: Optional[torch.Tensor] = None,
target_mapping=None, target_mapping: Optional[torch.Tensor] = None,
token_type_ids=None, token_type_ids: Optional[torch.Tensor] = None,
input_mask=None, input_mask: Optional[torch.Tensor] = None,
head_mask=None, head_mask: Optional[torch.Tensor] = None,
inputs_embeds=None, inputs_embeds: Optional[torch.Tensor] = None,
start_positions=None, start_positions: Optional[torch.Tensor] = None,
end_positions=None, end_positions: Optional[torch.Tensor] = None,
is_impossible=None, is_impossible: Optional[torch.Tensor] = None,
cls_index=None, cls_index: Optional[torch.Tensor] = None,
p_mask=None, p_mask: Optional[torch.Tensor] = None,
use_mems=None, use_mems: Optional[bool] = None,
output_attentions=None, output_attentions: Optional[bool] = None,
output_hidden_states=None, output_hidden_states: Optional[bool] = None,
return_dict=None, return_dict: Optional[bool] = None,
**kwargs, # delete when `use_cache` is removed in XLNetModel **kwargs, # delete when `use_cache` is removed in XLNetModel
): ) -> Union[Tuple, XLNetForQuestionAnsweringOutput]:
r""" r"""
start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
Labels for position (index) of the start of the labelled span for computing the token classification loss. Labels for position (index) of the start of the labelled span for computing the token classification loss.
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment