Type hints added for TFMobileBert (#16505)

* Type hints added

* make style

* Return type hints added

* fixed typo
Co-authored-by: matt <rocketknight1@gmail.com>

src/transformers/models/mobilebert/modeling_tf_mobilebert.py

@@ -17,8 +17,9 @@
 
 import warnings
 from dataclasses import dataclass
-from typing import Dict, Optional, Tuple
+from typing import Dict, Optional, Tuple, Union
 
+import numpy as np
 import tensorflow as tf
 
 from ...activations_tf import get_tf_activation
@@ -34,6 +35,7 @@ from ...modeling_tf_outputs import (
 )
 from ...modeling_tf_utils import (
     TFMaskedLanguageModelingLoss,
+    TFModelInputType,
     TFMultipleChoiceLoss,
     TFNextSentencePredictionLoss,
     TFPreTrainedModel,
@@ -934,17 +936,17 @@ class TFMobileBertModel(TFMobileBertPreTrainedModel):
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFBaseModelOutputWithPooling]:
         outputs = self.mobilebert(
             input_ids=input_ids,
             attention_mask=attention_mask,
@@ -998,17 +1000,17 @@ class TFMobileBertForPreTraining(TFMobileBertPreTrainedModel):
     @replace_return_docstrings(output_type=TFMobileBertForPreTrainingOutput, config_class=_CONFIG_FOR_DOC)
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFMobileBertForPreTrainingOutput]:
         r"""
         Return:
 
@@ -1097,18 +1099,18 @@ class TFMobileBertForMaskedLM(TFMobileBertPreTrainedModel, TFMaskedLanguageModel
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFMaskedLMOutput]:
         r"""
         labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
             Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
@@ -1180,18 +1182,18 @@ class TFMobileBertForNextSentencePrediction(TFMobileBertPreTrainedModel, TFNextS
     @replace_return_docstrings(output_type=TFNextSentencePredictorOutput, config_class=_CONFIG_FOR_DOC)
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        next_sentence_label=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        next_sentence_label: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFNextSentencePredictorOutput]:
         r"""
         Return:
 
@@ -1292,18 +1294,18 @@ class TFMobileBertForSequenceClassification(TFMobileBertPreTrainedModel, TFSeque
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFSequenceClassifierOutput]:
         r"""
         labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
             Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
@@ -1388,19 +1390,19 @@ class TFMobileBertForQuestionAnswering(TFMobileBertPreTrainedModel, TFQuestionAn
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        start_positions=None,
-        end_positions=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        start_positions: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        end_positions: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFQuestionAnsweringModelOutput]:
         r"""
         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.
@@ -1505,18 +1507,18 @@ class TFMobileBertForMultipleChoice(TFMobileBertPreTrainedModel, TFMultipleChoic
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFMultipleChoiceModelOutput]:
         r"""
         labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
             Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]`
@@ -1634,18 +1636,18 @@ class TFMobileBertForTokenClassification(TFMobileBertPreTrainedModel, TFTokenCla
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
-    ):
+        input_ids: Optional[TFModelInputType] = None,
+        attention_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        token_type_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        position_ids: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        head_mask: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        inputs_embeds: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[Union[np.ndarray, tf.Tensor]] = None,
+        training: Optional[bool] = False,
+    ) -> Union[Tuple, TFTokenClassifierOutput]:
         r"""
         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]`.