Add DeBERTa head models (#9691)

* Add DebertaForMaskedLM, DebertaForTokenClassification, DebertaForQuestionAnswering

* Add docs and fix quality

* Fix Deberta not having pooler

docs/source/model_doc/deberta.rst

@@ -70,8 +70,29 @@ DebertaPreTrainedModel
     :members:
 
 
+DebertaForMaskedLM
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DebertaForMaskedLM
+    :members:
+
+
 DebertaForSequenceClassification
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: transformers.DebertaForSequenceClassification
     :members:
+
+
+DebertaForTokenClassification
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DebertaForTokenClassification
+    :members:
+
+
+DebertaForQuestionAnswering
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: transformers.DebertaForQuestionAnswering
+    :members:

src/transformers/__init__.py

@@ -477,7 +477,10 @@ if is_torch_available():
             "DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
             "DebertaForSequenceClassification",
             "DebertaModel",
+            "DebertaForMaskedLM",
             "DebertaPreTrainedModel",
+            "DebertaForTokenClassification",
+            "DebertaForQuestionAnswering",
         ]
     )
     _import_structure["models.distilbert"].extend(
@@ -1527,7 +1530,10 @@ if TYPE_CHECKING:
         )
         from .models.deberta import (
             DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
+            DebertaForMaskedLM,
+            DebertaForQuestionAnswering,
             DebertaForSequenceClassification,
+            DebertaForTokenClassification,
             DebertaModel,
             DebertaPreTrainedModel,
         )

src/transformers/models/auto/modeling_auto.py

@@ -62,7 +62,13 @@ from ..camembert.modeling_camembert import (
     CamembertModel,
 )
 from ..ctrl.modeling_ctrl import CTRLForSequenceClassification, CTRLLMHeadModel, CTRLModel
-from ..deberta.modeling_deberta import DebertaForSequenceClassification, DebertaModel
+from ..deberta.modeling_deberta import (
+    DebertaForMaskedLM,
+    DebertaForQuestionAnswering,
+    DebertaForSequenceClassification,
+    DebertaForTokenClassification,
+    DebertaModel,
+)
 from ..distilbert.modeling_distilbert import (
     DistilBertForMaskedLM,
     DistilBertForMultipleChoice,
@@ -378,6 +384,7 @@ MODEL_WITH_LM_HEAD_MAPPING = OrderedDict(
         (FunnelConfig, FunnelForMaskedLM),
         (MPNetConfig, MPNetForMaskedLM),
         (TapasConfig, TapasForMaskedLM),
+        (DebertaConfig, DebertaForMaskedLM),
     ]
 )
 
@@ -426,6 +433,7 @@ MODEL_FOR_MASKED_LM_MAPPING = OrderedDict(
         (FunnelConfig, FunnelForMaskedLM),
         (MPNetConfig, MPNetForMaskedLM),
         (TapasConfig, TapasForMaskedLM),
+        (DebertaConfig, DebertaForMaskedLM),
     ]
 )
 
@@ -503,6 +511,7 @@ MODEL_FOR_QUESTION_ANSWERING_MAPPING = OrderedDict(
         (FunnelConfig, FunnelForQuestionAnswering),
         (LxmertConfig, LxmertForQuestionAnswering),
         (MPNetConfig, MPNetForQuestionAnswering),
+        (DebertaConfig, DebertaForQuestionAnswering),
     ]
 )
 
@@ -533,6 +542,7 @@ MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING = OrderedDict(
         (FlaubertConfig, FlaubertForTokenClassification),
         (FunnelConfig, FunnelForTokenClassification),
         (MPNetConfig, MPNetForTokenClassification),
+        (DebertaConfig, DebertaForTokenClassification),
     ]
 )
 

src/transformers/models/deberta/__init__.py

@@ -31,7 +31,10 @@ if is_torch_available():
         "DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
         "DebertaForSequenceClassification",
         "DebertaModel",
+        "DebertaForMaskedLM",
         "DebertaPreTrainedModel",
+        "DebertaForTokenClassification",
+        "DebertaForQuestionAnswering",
     ]
 
 
@@ -42,7 +45,10 @@ if TYPE_CHECKING:
     if is_torch_available():
         from .modeling_deberta import (
             DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
+            DebertaForMaskedLM,
+            DebertaForQuestionAnswering,
             DebertaForSequenceClassification,
+            DebertaForTokenClassification,
             DebertaModel,
             DebertaPreTrainedModel,
         )

src/transformers/models/deberta/modeling_deberta.py

@@ -24,7 +24,13 @@ from torch.nn import CrossEntropyLoss
 
 from ...activations import ACT2FN
 from ...file_utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward
-from ...modeling_outputs import BaseModelOutput, SequenceClassifierOutput
+from ...modeling_outputs import (
+    BaseModelOutput,
+    MaskedLMOutput,
+    QuestionAnsweringModelOutput,
+    SequenceClassifierOutput,
+    TokenClassifierOutput,
+)
 from ...modeling_utils import PreTrainedModel
 from ...utils import logging
 from .configuration_deberta import DebertaConfig
@@ -945,6 +951,135 @@ class DebertaModel(DebertaPreTrainedModel):
         )
 
 
+@add_start_docstrings("""DeBERTa Model with a `language modeling` head on top. """, DEBERTA_START_DOCSTRING)
+class DebertaForMaskedLM(DebertaPreTrainedModel):
+
+    _keys_to_ignore_on_load_unexpected = [r"pooler"]
+    _keys_to_ignore_on_load_missing = [r"position_ids", r"predictions.decoder.bias"]
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.deberta = DebertaModel(config)
+        self.cls = DebertaOnlyMLMHead(config)
+
+        self.init_weights()
+
+    def get_output_embeddings(self):
+        return self.cls.predictions.decoder
+
+    def set_output_embeddings(self, new_embeddings):
+        self.cls.predictions.decoder = new_embeddings
+
+    @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        tokenizer_class=_TOKENIZER_FOR_DOC,
+        checkpoint="microsoft/deberta-base",
+        output_type=MaskedLMOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        inputs_embeds=None,
+        labels=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Labels for computing the masked language modeling loss. Indices should be in ``[-100, 0, ...,
+            config.vocab_size]`` (see ``input_ids`` docstring) Tokens with indices set to ``-100`` are ignored
+            (masked), the loss is only computed for the tokens with labels in ``[0, ..., config.vocab_size]``
+        """
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.deberta(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+        prediction_scores = self.cls(sequence_output)
+
+        masked_lm_loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()  # -100 index = padding token
+            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+        if not return_dict:
+            output = (prediction_scores,) + outputs[1:]
+            return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
+
+        return MaskedLMOutput(
+            loss=masked_lm_loss,
+            logits=prediction_scores,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+# copied from transformers.models.bert.BertPredictionHeadTransform with bert -> deberta
+class DebertaPredictionHeadTransform(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+        if isinstance(config.hidden_act, str):
+            self.transform_act_fn = ACT2FN[config.hidden_act]
+        else:
+            self.transform_act_fn = config.hidden_act
+        self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+    def forward(self, hidden_states):
+        hidden_states = self.dense(hidden_states)
+        hidden_states = self.transform_act_fn(hidden_states)
+        hidden_states = self.LayerNorm(hidden_states)
+        return hidden_states
+
+
+# copied from transformers.models.bert.BertLMPredictionHead with bert -> deberta
+class DebertaLMPredictionHead(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.transform = DebertaPredictionHeadTransform(config)
+
+        # The output weights are the same as the input embeddings, but there is
+        # an output-only bias for each token.
+        self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        self.bias = nn.Parameter(torch.zeros(config.vocab_size))
+
+        # Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
+        self.decoder.bias = self.bias
+
+    def forward(self, hidden_states):
+        hidden_states = self.transform(hidden_states)
+        hidden_states = self.decoder(hidden_states)
+        return hidden_states
+
+
+# copied from transformers.models.bert.BertOnlyMLMHead with bert -> deberta
+class DebertaOnlyMLMHead(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.predictions = DebertaLMPredictionHead(config)
+
+    def forward(self, sequence_output):
+        prediction_scores = self.predictions(sequence_output)
+        return prediction_scores
+
+
 @add_start_docstrings(
     """
     DeBERTa Model transformer with a sequence classification/regression head on top (a linear layer on top of the
@@ -1049,3 +1184,192 @@ class DebertaForSequenceClassification(DebertaPreTrainedModel):
                 hidden_states=outputs.hidden_states,
                 attentions=outputs.attentions,
             )
+
+
+@add_start_docstrings(
+    """
+    DeBERTa Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for
+    Named-Entity-Recognition (NER) tasks.
+    """,
+    DEBERTA_START_DOCSTRING,
+)
+class DebertaForTokenClassification(DebertaPreTrainedModel):
+
+    _keys_to_ignore_on_load_unexpected = [r"pooler"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.deberta = DebertaModel(config)
+        self.dropout = nn.Dropout(config.hidden_dropout_prob)
+        self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+    @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        tokenizer_class=_TOKENIZER_FOR_DOC,
+        checkpoint="microsoft/deberta-base",
+        output_type=TokenClassifierOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        inputs_embeds=None,
+        labels=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        labels (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
+            Labels for computing the token classification loss. Indices should be in ``[0, ..., config.num_labels -
+            1]``.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.deberta(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+
+        sequence_output = self.dropout(sequence_output)
+        logits = self.classifier(sequence_output)
+
+        loss = None
+        if labels is not None:
+            loss_fct = CrossEntropyLoss()
+            # Only keep active parts of the loss
+            if attention_mask is not None:
+                active_loss = attention_mask.view(-1) == 1
+                active_logits = logits.view(-1, self.num_labels)
+                active_labels = torch.where(
+                    active_loss, labels.view(-1), torch.tensor(loss_fct.ignore_index).type_as(labels)
+                )
+                loss = loss_fct(active_logits, active_labels)
+            else:
+                loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return TokenClassifierOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """
+    DeBERTa Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear
+    layers on top of the hidden-states output to compute `span start logits` and `span end logits`).
+    """,
+    DEBERTA_START_DOCSTRING,
+)
+class DebertaForQuestionAnswering(DebertaPreTrainedModel):
+
+    _keys_to_ignore_on_load_unexpected = [r"pooler"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+
+        self.deberta = DebertaModel(config)
+        self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
+
+        self.init_weights()
+
+    @add_start_docstrings_to_model_forward(DEBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+    @add_code_sample_docstrings(
+        tokenizer_class=_TOKENIZER_FOR_DOC,
+        checkpoint="microsoft/deberta-base",
+        output_type=QuestionAnsweringModelOutput,
+        config_class=_CONFIG_FOR_DOC,
+    )
+    def forward(
+        self,
+        input_ids=None,
+        attention_mask=None,
+        token_type_ids=None,
+        position_ids=None,
+        inputs_embeds=None,
+        start_positions=None,
+        end_positions=None,
+        output_attentions=None,
+        output_hidden_states=None,
+        return_dict=None,
+    ):
+        r"""
+        start_positions (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`):
+            Labels for position (index) of the start of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (:obj:`sequence_length`). Position outside of the
+            sequence are not taken into account for computing the loss.
+        end_positions (:obj:`torch.LongTensor` of shape :obj:`(batch_size,)`, `optional`):
+            Labels for position (index) of the end of the labelled span for computing the token classification loss.
+            Positions are clamped to the length of the sequence (:obj:`sequence_length`). Position outside of the
+            sequence are not taken into account for computing the loss.
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        outputs = self.deberta(
+            input_ids,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        sequence_output = outputs[0]
+
+        logits = self.qa_outputs(sequence_output)
+        start_logits, end_logits = logits.split(1, dim=-1)
+        start_logits = start_logits.squeeze(-1)
+        end_logits = end_logits.squeeze(-1)
+
+        total_loss = None
+        if start_positions is not None and end_positions is not None:
+            # If we are on multi-GPU, split add a dimension
+            if len(start_positions.size()) > 1:
+                start_positions = start_positions.squeeze(-1)
+            if len(end_positions.size()) > 1:
+                end_positions = end_positions.squeeze(-1)
+            # sometimes the start/end positions are outside our model inputs, we ignore these terms
+            ignored_index = start_logits.size(1)
+            start_positions.clamp_(0, ignored_index)
+            end_positions.clamp_(0, ignored_index)
+
+            loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
+            start_loss = loss_fct(start_logits, start_positions)
+            end_loss = loss_fct(end_logits, end_positions)
+            total_loss = (start_loss + end_loss) / 2
+
+        if not return_dict:
+            output = (start_logits, end_logits) + outputs[1:]
+            return ((total_loss,) + output) if total_loss is not None else output
+
+        return QuestionAnsweringModelOutput(
+            loss=total_loss,
+            start_logits=start_logits,
+            end_logits=end_logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )

src/transformers/utils/dummy_pt_objects.py

@@ -739,6 +739,24 @@ class CTRLPreTrainedModel:
 DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST = None
 
 
+class DebertaForMaskedLM:
+    def __init__(self, *args, **kwargs):
+        requires_pytorch(self)
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_pytorch(self)
+
+
+class DebertaForQuestionAnswering:
+    def __init__(self, *args, **kwargs):
+        requires_pytorch(self)
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_pytorch(self)
+
+
 class DebertaForSequenceClassification:
     def __init__(self, *args, **kwargs):
         requires_pytorch(self)
@@ -748,6 +766,15 @@ class DebertaForSequenceClassification:
         requires_pytorch(self)
 
 
+class DebertaForTokenClassification:
+    def __init__(self, *args, **kwargs):
+        requires_pytorch(self)
+
+    @classmethod
+    def from_pretrained(self, *args, **kwargs):
+        requires_pytorch(self)
+
+
 class DebertaModel:
     def __init__(self, *args, **kwargs):
         requires_pytorch(self)

tests/test_modeling_deberta.py

@@ -1,251 +1,290 @@
-# coding=utf-8
+# coding=utf-8
-# Copyright 2018 Microsoft Authors and the HuggingFace Inc. team.
+# Copyright 2018 Microsoft Authors and the HuggingFace Inc. team.
-#
+#
-# Licensed under the Apache License, Version 2.0 (the "License");
+# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
+# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
+# You may obtain a copy of the License at
-#
+#
-#     http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
-#
+#
-# Unless required by applicable law or agreed to in writing, software
+# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
+# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
+# See the License for the specific language governing permissions and
-# limitations under the License.
+# limitations under the License.
-
+
-
+
-import random
+import random
-import unittest
+import unittest
-
+
-import numpy as np
+import numpy as np
-
+
-from transformers import is_torch_available
+from transformers import is_torch_available
-from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device
+from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device
-
+
-from .test_configuration_common import ConfigTester
+from .test_configuration_common import ConfigTester
-from .test_modeling_common import ModelTesterMixin, ids_tensor
+from .test_modeling_common import ModelTesterMixin, ids_tensor
-
+
-
+
-if is_torch_available():
+if is_torch_available():
-    import torch
+    import torch
-
+
-    from transformers import (  # XxxForMaskedLM,; XxxForQuestionAnswering,; XxxForTokenClassification,
+    from transformers import (
-        DebertaConfig,
+        DebertaConfig,
-        DebertaForSequenceClassification,
+        DebertaForMaskedLM,
-        DebertaModel,
+        DebertaForQuestionAnswering,
-    )
+        DebertaForSequenceClassification,
-    from transformers.models.deberta.modeling_deberta import DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST
+        DebertaForTokenClassification,
-
+        DebertaModel,
-
+    )
-@require_torch
+    from transformers.models.deberta.modeling_deberta import DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST
-class DebertaModelTest(ModelTesterMixin, unittest.TestCase):
+
-
+
-    all_model_classes = (
+@require_torch
-        (
+class DebertaModelTest(ModelTesterMixin, unittest.TestCase):
-            DebertaModel,
+
-            DebertaForSequenceClassification,
+    all_model_classes = (
-        )  # , DebertaForMaskedLM, DebertaForQuestionAnswering, DebertaForTokenClassification)
+        (
-        if is_torch_available()
+            DebertaModel,
-        else ()
+            DebertaForMaskedLM,
-    )
+            DebertaForSequenceClassification,
-
+            DebertaForTokenClassification,
-    test_torchscript = False
+            DebertaForQuestionAnswering,
-    test_pruning = False
+        )
-    test_head_masking = False
+        if is_torch_available()
-    is_encoder_decoder = False
+        else ()
-
+    )
-    class DebertaModelTester(object):
+
-        def __init__(
+    test_torchscript = False
-            self,
+    test_pruning = False
-            parent,
+    test_head_masking = False
-            batch_size=13,
+    is_encoder_decoder = False
-            seq_length=7,
+
-            is_training=True,
+    class DebertaModelTester(object):
-            use_input_mask=True,
+        def __init__(
-            use_token_type_ids=True,
+            self,
-            use_labels=True,
+            parent,
-            vocab_size=99,
+            batch_size=13,
-            hidden_size=32,
+            seq_length=7,
-            num_hidden_layers=5,
+            is_training=True,
-            num_attention_heads=4,
+            use_input_mask=True,
-            intermediate_size=37,
+            use_token_type_ids=True,
-            hidden_act="gelu",
+            use_labels=True,
-            hidden_dropout_prob=0.1,
+            vocab_size=99,
-            attention_probs_dropout_prob=0.1,
+            hidden_size=32,
-            max_position_embeddings=512,
+            num_hidden_layers=5,
-            type_vocab_size=16,
+            num_attention_heads=4,
-            type_sequence_label_size=2,
+            intermediate_size=37,
-            initializer_range=0.02,
+            hidden_act="gelu",
-            relative_attention=False,
+            hidden_dropout_prob=0.1,
-            position_biased_input=True,
+            attention_probs_dropout_prob=0.1,
-            pos_att_type="None",
+            max_position_embeddings=512,
-            num_labels=3,
+            type_vocab_size=16,
-            num_choices=4,
+            type_sequence_label_size=2,
-            scope=None,
+            initializer_range=0.02,
-        ):
+            relative_attention=False,
-            self.parent = parent
+            position_biased_input=True,
-            self.batch_size = batch_size
+            pos_att_type="None",
-            self.seq_length = seq_length
+            num_labels=3,
-            self.is_training = is_training
+            num_choices=4,
-            self.use_input_mask = use_input_mask
+            scope=None,
-            self.use_token_type_ids = use_token_type_ids
+        ):
-            self.use_labels = use_labels
+            self.parent = parent
-            self.vocab_size = vocab_size
+            self.batch_size = batch_size
-            self.hidden_size = hidden_size
+            self.seq_length = seq_length
-            self.num_hidden_layers = num_hidden_layers
+            self.is_training = is_training
-            self.num_attention_heads = num_attention_heads
+            self.use_input_mask = use_input_mask
-            self.intermediate_size = intermediate_size
+            self.use_token_type_ids = use_token_type_ids
-            self.hidden_act = hidden_act
+            self.use_labels = use_labels
-            self.hidden_dropout_prob = hidden_dropout_prob
+            self.vocab_size = vocab_size
-            self.attention_probs_dropout_prob = attention_probs_dropout_prob
+            self.hidden_size = hidden_size
-            self.max_position_embeddings = max_position_embeddings
+            self.num_hidden_layers = num_hidden_layers
-            self.type_vocab_size = type_vocab_size
+            self.num_attention_heads = num_attention_heads
-            self.type_sequence_label_size = type_sequence_label_size
+            self.intermediate_size = intermediate_size
-            self.initializer_range = initializer_range
+            self.hidden_act = hidden_act
-            self.num_labels = num_labels
+            self.hidden_dropout_prob = hidden_dropout_prob
-            self.num_choices = num_choices
+            self.attention_probs_dropout_prob = attention_probs_dropout_prob
-            self.relative_attention = relative_attention
+            self.max_position_embeddings = max_position_embeddings
-            self.position_biased_input = position_biased_input
+            self.type_vocab_size = type_vocab_size
-            self.pos_att_type = pos_att_type
+            self.type_sequence_label_size = type_sequence_label_size
-            self.scope = scope
+            self.initializer_range = initializer_range
-
+            self.num_labels = num_labels
-        def prepare_config_and_inputs(self):
+            self.num_choices = num_choices
-            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
+            self.relative_attention = relative_attention
-
+            self.position_biased_input = position_biased_input
-            input_mask = None
+            self.pos_att_type = pos_att_type
-            if self.use_input_mask:
+            self.scope = scope
-                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
+
-
+        def prepare_config_and_inputs(self):
-            token_type_ids = None
+            input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)
-            if self.use_token_type_ids:
+
-                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
+            input_mask = None
-
+            if self.use_input_mask:
-            sequence_labels = None
+                input_mask = ids_tensor([self.batch_size, self.seq_length], vocab_size=2)
-            token_labels = None
+
-            choice_labels = None
+            token_type_ids = None
-            if self.use_labels:
+            if self.use_token_type_ids:
-                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
+                token_type_ids = ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size)
-                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
+
-                choice_labels = ids_tensor([self.batch_size], self.num_choices)
+            sequence_labels = None
-
+            token_labels = None
-            config = DebertaConfig(
+            choice_labels = None
-                vocab_size=self.vocab_size,
+            if self.use_labels:
-                hidden_size=self.hidden_size,
+                sequence_labels = ids_tensor([self.batch_size], self.type_sequence_label_size)
-                num_hidden_layers=self.num_hidden_layers,
+                token_labels = ids_tensor([self.batch_size, self.seq_length], self.num_labels)
-                num_attention_heads=self.num_attention_heads,
+                choice_labels = ids_tensor([self.batch_size], self.num_choices)
-                intermediate_size=self.intermediate_size,
+
-                hidden_act=self.hidden_act,
+            config = DebertaConfig(
-                hidden_dropout_prob=self.hidden_dropout_prob,
+                vocab_size=self.vocab_size,
-                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
+                hidden_size=self.hidden_size,
-                max_position_embeddings=self.max_position_embeddings,
+                num_hidden_layers=self.num_hidden_layers,
-                type_vocab_size=self.type_vocab_size,
+                num_attention_heads=self.num_attention_heads,
-                initializer_range=self.initializer_range,
+                intermediate_size=self.intermediate_size,
-                relative_attention=self.relative_attention,
+                hidden_act=self.hidden_act,
-                position_biased_input=self.position_biased_input,
+                hidden_dropout_prob=self.hidden_dropout_prob,
-                pos_att_type=self.pos_att_type,
+                attention_probs_dropout_prob=self.attention_probs_dropout_prob,
-            )
+                max_position_embeddings=self.max_position_embeddings,
-
+                type_vocab_size=self.type_vocab_size,
-            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+                initializer_range=self.initializer_range,
-
+                relative_attention=self.relative_attention,
-        def check_loss_output(self, result):
+                position_biased_input=self.position_biased_input,
-            self.parent.assertListEqual(list(result.loss.size()), [])
+                pos_att_type=self.pos_att_type,
-
+            )
-        def create_and_check_deberta_model(
+
-            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+            return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-        ):
+
-            model = DebertaModel(config=config)
+        def check_loss_output(self, result):
-            model.to(torch_device)
+            self.parent.assertListEqual(list(result.loss.size()), [])
-            model.eval()
+
-            sequence_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)[0]
+        def create_and_check_deberta_model(
-            sequence_output = model(input_ids, token_type_ids=token_type_ids)[0]
+            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-            sequence_output = model(input_ids)[0]
+        ):
-
+            model = DebertaModel(config=config)
-            self.parent.assertListEqual(
+            model.to(torch_device)
-                list(sequence_output.size()), [self.batch_size, self.seq_length, self.hidden_size]
+            model.eval()
-            )
+            sequence_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)[0]
-
+            sequence_output = model(input_ids, token_type_ids=token_type_ids)[0]
-        def create_and_check_deberta_for_sequence_classification(
+            sequence_output = model(input_ids)[0]
-            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
+
-        ):
+            self.parent.assertListEqual(
-            config.num_labels = self.num_labels
+                list(sequence_output.size()), [self.batch_size, self.seq_length, self.hidden_size]
-            model = DebertaForSequenceClassification(config)
+            )
-            model.to(torch_device)
+
-            model.eval()
+        def create_and_check_deberta_for_masked_lm(
-            result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
+            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-            self.parent.assertListEqual(list(result.logits.size()), [self.batch_size, self.num_labels])
+        ):
-            self.check_loss_output(result)
+            model = DebertaForMaskedLM(config=config)
-
+            model.to(torch_device)
-        def prepare_config_and_inputs_for_common(self):
+            model.eval()
-            config_and_inputs = self.prepare_config_and_inputs()
+            result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=token_labels)
-            (
+
-                config,
+            self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.vocab_size))
-                input_ids,
+
-                token_type_ids,
+        def create_and_check_deberta_for_sequence_classification(
-                input_mask,
+            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-                sequence_labels,
+        ):
-                token_labels,
+            config.num_labels = self.num_labels
-                choice_labels,
+            model = DebertaForSequenceClassification(config)
-            ) = config_and_inputs
+            model.to(torch_device)
-            inputs_dict = {"input_ids": input_ids, "token_type_ids": token_type_ids, "attention_mask": input_mask}
+            model.eval()
-            return config, inputs_dict
+            result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
-
+            self.parent.assertListEqual(list(result.logits.size()), [self.batch_size, self.num_labels])
-    def setUp(self):
+            self.check_loss_output(result)
-        self.model_tester = DebertaModelTest.DebertaModelTester(self)
+
-        self.config_tester = ConfigTester(self, config_class=DebertaConfig, hidden_size=37)
+        def create_and_check_deberta_for_token_classification(
-
+            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-    def test_config(self):
+        ):
-        self.config_tester.run_common_tests()
+            config.num_labels = self.num_labels
-
+            model = DebertaForTokenClassification(config=config)
-    def test_deberta_model(self):
+            model.to(torch_device)
-        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+            model.eval()
-        self.model_tester.create_and_check_deberta_model(*config_and_inputs)
+            result = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=token_labels)
-
+            self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.num_labels))
-    def test_for_sequence_classification(self):
+
-        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        def create_and_check_deberta_for_question_answering(
-        self.model_tester.create_and_check_deberta_for_sequence_classification(*config_and_inputs)
+            self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
-
+        ):
-    @unittest.skip(reason="Model not available yet")
+            model = DebertaForQuestionAnswering(config=config)
-    def test_for_masked_lm(self):
+            model.to(torch_device)
-        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+            model.eval()
-        self.model_tester.create_and_check_deberta_for_masked_lm(*config_and_inputs)
+            result = model(
-
+                input_ids,
-    @unittest.skip(reason="Model not available yet")
+                attention_mask=input_mask,
-    def test_for_question_answering(self):
+                token_type_ids=token_type_ids,
-        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+                start_positions=sequence_labels,
-        self.model_tester.create_and_check_deberta_for_question_answering(*config_and_inputs)
+                end_positions=sequence_labels,
-
+            )
-    @unittest.skip(reason="Model not available yet")
+            self.parent.assertEqual(result.start_logits.shape, (self.batch_size, self.seq_length))
-    def test_for_token_classification(self):
+            self.parent.assertEqual(result.end_logits.shape, (self.batch_size, self.seq_length))
-        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+
-        self.model_tester.create_and_check_deberta_for_token_classification(*config_and_inputs)
+        def prepare_config_and_inputs_for_common(self):
-
+            config_and_inputs = self.prepare_config_and_inputs()
-    @slow
+            (
-    def test_model_from_pretrained(self):
+                config,
-        for model_name in DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
+                input_ids,
-            model = DebertaModel.from_pretrained(model_name)
+                token_type_ids,
-            self.assertIsNotNone(model)
+                input_mask,
-
+                sequence_labels,
-
+                token_labels,
-@require_torch
+                choice_labels,
-@require_sentencepiece
+            ) = config_and_inputs
-@require_tokenizers
+            inputs_dict = {"input_ids": input_ids, "token_type_ids": token_type_ids, "attention_mask": input_mask}
-class DebertaModelIntegrationTest(unittest.TestCase):
+            return config, inputs_dict
-    @unittest.skip(reason="Model not available yet")
+
-    def test_inference_masked_lm(self):
+    def setUp(self):
-        pass
+        self.model_tester = DebertaModelTest.DebertaModelTester(self)
-
+        self.config_tester = ConfigTester(self, config_class=DebertaConfig, hidden_size=37)
-    @slow
+
-    def test_inference_no_head(self):
+    def test_config(self):
-        random.seed(0)
+        self.config_tester.run_common_tests()
-        np.random.seed(0)
+
-        torch.manual_seed(0)
+    def test_deberta_model(self):
-        torch.cuda.manual_seed_all(0)
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
-        model = DebertaModel.from_pretrained("microsoft/deberta-base")
+        self.model_tester.create_and_check_deberta_model(*config_and_inputs)
-
+
-        input_ids = torch.tensor([[0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2]])
+    def test_for_sequence_classification(self):
-        output = model(input_ids)[0]
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
-        # compare the actual values for a slice.
+        self.model_tester.create_and_check_deberta_for_sequence_classification(*config_and_inputs)
-        expected_slice = torch.tensor(
+
-            [[[-0.0218, -0.6641, -0.3665], [-0.3907, -0.4716, -0.6640], [0.7461, 1.2570, -0.9063]]]
+    def test_for_masked_lm(self):
-        )
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
-        self.assertTrue(torch.allclose(output[:, :3, :3], expected_slice, atol=1e-4), f"{output[:, :3, :3]}")
+        self.model_tester.create_and_check_deberta_for_masked_lm(*config_and_inputs)
+
+    def test_for_question_answering(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_deberta_for_question_answering(*config_and_inputs)
+
+    def test_for_token_classification(self):
+        config_and_inputs = self.model_tester.prepare_config_and_inputs()
+        self.model_tester.create_and_check_deberta_for_token_classification(*config_and_inputs)
+
+    @slow
+    def test_model_from_pretrained(self):
+        for model_name in DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
+            model = DebertaModel.from_pretrained(model_name)
+            self.assertIsNotNone(model)
+
+
+@require_torch
+@require_sentencepiece
+@require_tokenizers
+class DebertaModelIntegrationTest(unittest.TestCase):
+    @unittest.skip(reason="Model not available yet")
+    def test_inference_masked_lm(self):
+        pass
+
+    @slow
+    def test_inference_no_head(self):
+        random.seed(0)
+        np.random.seed(0)
+        torch.manual_seed(0)
+        torch.cuda.manual_seed_all(0)
+        model = DebertaModel.from_pretrained("microsoft/deberta-base")
+
+        input_ids = torch.tensor([[0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2]])
+        output = model(input_ids)[0]
+        # compare the actual values for a slice.
+        expected_slice = torch.tensor(
+            [[[-0.0218, -0.6641, -0.3665], [-0.3907, -0.4716, -0.6640], [0.7461, 1.2570, -0.9063]]]
+        )
+        self.assertTrue(torch.allclose(output[:, :3, :3], expected_slice, atol=1e-4), f"{output[:, :3, :3]}")