[Docs] More general docstrings (#14028)

* up

* finish

* up

* up

* finish

src/transformers/models/deberta_v2/modeling_deberta_v2.py

@@ -974,7 +974,7 @@ class DebertaV2Model(DebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1081,7 +1081,7 @@ class DebertaV2ForMaskedLM(DebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1222,7 +1222,7 @@ class DebertaV2ForSequenceClassification(DebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1318,7 +1318,7 @@ class DebertaV2ForTokenClassification(DebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1406,7 +1406,7 @@ class DebertaV2ForQuestionAnswering(DebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/deberta_v2/modeling_tf_deberta_v2.py

@@ -1223,7 +1223,7 @@ class TFDebertaV2Model(TFDebertaV2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFBaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1296,7 +1296,7 @@ class TFDebertaV2ForMaskedLM(TFDebertaV2PreTrainedModel, TFMaskedLanguageModelin
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFMaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1399,7 +1399,7 @@ class TFDebertaV2ForSequenceClassification(TFDebertaV2PreTrainedModel, TFSequenc
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1497,7 +1497,7 @@ class TFDebertaV2ForTokenClassification(TFDebertaV2PreTrainedModel, TFTokenClass
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1591,7 +1591,7 @@ class TFDebertaV2ForQuestionAnswering(TFDebertaV2PreTrainedModel, TFQuestionAnsw
 
     @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/distilbert/modeling_distilbert.py

@@ -508,7 +508,7 @@ class DistilBertModel(DistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, num_choices"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -604,7 +604,7 @@ class DistilBertForMaskedLM(DistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, num_choices"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -702,7 +702,7 @@ class DistilBertForSequenceClassification(DistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -818,7 +818,7 @@ class DistilBertForQuestionAnswering(DistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, num_choices"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -935,7 +935,7 @@ class DistilBertForTokenClassification(DistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/distilbert/modeling_tf_distilbert.py

@@ -543,7 +543,7 @@ class TFDistilBertModel(TFDistilBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFBaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -658,7 +658,7 @@ class TFDistilBertForMaskedLM(TFDistilBertPreTrainedModel, TFMaskedLanguageModel
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFMaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -759,7 +759,7 @@ class TFDistilBertForSequenceClassification(TFDistilBertPreTrainedModel, TFSeque
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -854,7 +854,7 @@ class TFDistilBertForTokenClassification(TFDistilBertPreTrainedModel, TFTokenCla
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -962,7 +962,7 @@ class TFDistilBertForMultipleChoice(TFDistilBertPreTrainedModel, TFMultipleChoic
         DISTILBERT_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length")
     )
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFMultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1088,7 +1088,7 @@ class TFDistilBertForQuestionAnswering(TFDistilBertPreTrainedModel, TFQuestionAn
 
     @add_start_docstrings_to_model_forward(DISTILBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/electra/modeling_electra.py

@@ -833,7 +833,7 @@ class ElectraModel(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutputWithCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -941,7 +941,7 @@ class ElectraForSequenceClassification(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1136,7 +1136,7 @@ class ElectraForMaskedLM(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1218,7 +1218,7 @@ class ElectraForTokenClassification(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1307,7 +1307,7 @@ class ElectraForQuestionAnswering(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1408,7 +1408,7 @@ class ElectraForMultipleChoice(ElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/electra/modeling_tf_electra.py

@@ -950,7 +950,7 @@ class TFElectraModel(TFElectraPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFBaseModelOutputWithPastAndCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -1211,7 +1211,7 @@ class TFElectraForMaskedLM(TFElectraPreTrainedModel, TFMaskedLanguageModelingLos
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFMaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1336,7 +1336,7 @@ class TFElectraForSequenceClassification(TFElectraPreTrainedModel, TFSequenceCla
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1444,7 +1444,7 @@ class TFElectraForMultipleChoice(TFElectraPreTrainedModel, TFMultipleChoiceLoss)
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFMultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1584,7 +1584,7 @@ class TFElectraForTokenClassification(TFElectraPreTrainedModel, TFTokenClassific
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1681,7 +1681,7 @@ class TFElectraForQuestionAnswering(TFElectraPreTrainedModel, TFQuestionAnswerin
 
     @add_start_docstrings_to_model_forward(ELECTRA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/flaubert/modeling_flaubert.py

@@ -148,7 +148,7 @@ class FlaubertModel(XLMModel):
 
     @add_start_docstrings_to_model_forward(FLAUBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/flaubert/modeling_tf_flaubert.py

@@ -236,7 +236,7 @@ class TFFlaubertModel(TFFlaubertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FLAUBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFBaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -820,7 +820,7 @@ class TFFlaubertWithLMHeadModel(TFFlaubertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FLAUBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFFlaubertWithLMHeadModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/fnet/modeling_fnet.py

@@ -545,7 +545,7 @@ class FNetModel(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -733,7 +733,7 @@ class FNetForMaskedLM(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -889,7 +889,7 @@ class FNetForSequenceClassification(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -961,7 +961,7 @@ class FNetForMultipleChoice(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1042,7 +1042,7 @@ class FNetForTokenClassification(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1111,7 +1111,7 @@ class FNetForQuestionAnswering(FNetPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FNET_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/fsmt/modeling_fsmt.py

@@ -1007,7 +1007,7 @@ class FSMTModel(PretrainedFSMTModel):
 
     @add_start_docstrings_to_model_forward(FSMT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/funnel/modeling_funnel.py

@@ -910,7 +910,7 @@ class FunnelBaseModel(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -987,7 +987,7 @@ class FunnelModel(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1174,7 +1174,7 @@ class FunnelForMaskedLM(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1248,7 +1248,7 @@ class FunnelForSequenceClassification(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1338,7 +1338,7 @@ class FunnelForMultipleChoice(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=MultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1424,7 +1424,7 @@ class FunnelForTokenClassification(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1506,7 +1506,7 @@ class FunnelForQuestionAnswering(FunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/funnel/modeling_tf_funnel.py

@@ -1126,7 +1126,7 @@ class TFFunnelBaseModel(TFFunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=TFBaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1187,7 +1187,7 @@ class TFFunnelModel(TFFunnelPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small",
         output_type=TFBaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1337,7 +1337,7 @@ class TFFunnelForMaskedLM(TFFunnelPreTrainedModel, TFMaskedLanguageModelingLoss)
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small",
         output_type=TFMaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1426,7 +1426,7 @@ class TFFunnelForSequenceClassification(TFFunnelPreTrainedModel, TFSequenceClass
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=TFSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1525,7 +1525,7 @@ class TFFunnelForMultipleChoice(TFFunnelPreTrainedModel, TFMultipleChoiceLoss):
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small-base",
         output_type=TFMultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1655,7 +1655,7 @@ class TFFunnelForTokenClassification(TFFunnelPreTrainedModel, TFTokenClassificat
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small",
         output_type=TFTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1747,7 +1747,7 @@ class TFFunnelForQuestionAnswering(TFFunnelPreTrainedModel, TFQuestionAnsweringL
 
     @add_start_docstrings_to_model_forward(FUNNEL_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="funnel-transformer/small",
         output_type=TFQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/gpt2/modeling_gpt2.py

@@ -732,7 +732,7 @@ class GPT2Model(GPT2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutputWithPastAndCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -1009,7 +1009,7 @@ class GPT2LMHeadModel(GPT2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutputWithCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -1338,7 +1338,7 @@ class GPT2ForSequenceClassification(GPT2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="microsoft/DialogRPT-updown",
         output_type=SequenceClassifierOutputWithPast,
         config_class=_CONFIG_FOR_DOC,
@@ -1457,7 +1457,7 @@ class GPT2ForTokenClassification(GPT2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="microsoft/DialogRPT-updown",
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/gpt2/modeling_tf_gpt2.py

@@ -587,7 +587,7 @@ class TFGPT2Model(TFGPT2PreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFBaseModelOutputWithPast,
         config_class=_CONFIG_FOR_DOC,
@@ -679,7 +679,7 @@ class TFGPT2LMHeadModel(TFGPT2PreTrainedModel, TFCausalLanguageModelingLoss):
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFCausalLMOutputWithPast,
         config_class=_CONFIG_FOR_DOC,
@@ -959,7 +959,7 @@ class TFGPT2ForSequenceClassification(TFGPT2PreTrainedModel, TFSequenceClassific
 
     @add_start_docstrings_to_model_forward(GPT2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="microsoft/DialogRPT-updown",
         output_type=TFSequenceClassifierOutputWithPast,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/gpt_neo/modeling_gpt_neo.py

@@ -497,7 +497,7 @@ class GPTNeoModel(GPTNeoPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT_NEO_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutputWithPastAndCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -713,7 +713,7 @@ class GPTNeoForCausalLM(GPTNeoPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT_NEO_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutputWithCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -827,7 +827,7 @@ class GPTNeoForSequenceClassification(GPTNeoPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPT_NEO_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutputWithPast,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/gptj/modeling_gptj.py

@@ -491,7 +491,7 @@ class GPTJModel(GPTJPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPTJ_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutputWithPast,
         config_class=_CONFIG_FOR_DOC,
@@ -743,7 +743,7 @@ class GPTJForCausalLM(GPTJPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPTJ_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutputWithPast,
         config_class=_CONFIG_FOR_DOC,
@@ -864,7 +864,7 @@ class GPTJForSequenceClassification(GPTJPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(GPTJ_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutputWithPast,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/ibert/modeling_ibert.py

@@ -772,7 +772,7 @@ class IBertModel(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutputWithPoolingAndCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
@@ -875,7 +875,7 @@ class IBertForMaskedLM(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MaskedLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -983,7 +983,7 @@ class IBertForSequenceClassification(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1066,7 +1066,7 @@ class IBertForMultipleChoice(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=MultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1160,7 +1160,7 @@ class IBertForTokenClassification(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1269,7 +1269,7 @@ class IBertForQuestionAnswering(IBertPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(IBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/led/modeling_led.py

@@ -2174,7 +2174,7 @@ class LEDModel(LEDPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LED_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2468,7 +2468,7 @@ class LEDForSequenceClassification(LEDPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LED_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2578,7 +2578,7 @@ class LEDForQuestionAnswering(LEDPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LED_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/led/modeling_tf_led.py

@@ -2230,7 +2230,7 @@ class TFLEDModel(TFLEDPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LED_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFLEDSeq2SeqModelOutput,
         config_class=_CONFIG_FOR_DOC,

src/transformers/models/longformer/modeling_longformer.py

@@ -1822,7 +1822,7 @@ class LongformerForSequenceClassification(LongformerPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=LongformerSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2084,7 +2084,7 @@ class LongformerForTokenClassification(LongformerPreTrainedModel):
 
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=LongformerTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2176,7 +2176,7 @@ class LongformerForMultipleChoice(LongformerPreTrainedModel):
         LONGFORMER_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length")
     )
     @add_code_sample_docstrings(
-        tokenizer_class=_TOKENIZER_FOR_DOC,
+        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=LongformerMultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,