From ceee95f763eb93aaad0491916aefd61dca0ee0a5 Mon Sep 17 00:00:00 2001
From: "zhangzhicheng.zzc" <zhangzhicheng.zzc@alibaba-inc.com>
Date: Sat, 6 Aug 2022 13:55:46 +0800
Subject: [PATCH] [to #43115513] bug fix for nlp backbone-head trainers Link:
 https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9662281

---
 modelscope/models/nlp/backbones/structbert.py | 33 +++++++++----------
 .../models/nlp/structbert/modeling_sbert.py   | 31 +++++++++--------
 2 files changed, 31 insertions(+), 33 deletions(-)

diff --git a/modelscope/models/nlp/backbones/structbert.py b/modelscope/models/nlp/backbones/structbert.py
index 125db040..cc062129 100644
--- a/modelscope/models/nlp/backbones/structbert.py
+++ b/modelscope/models/nlp/backbones/structbert.py
@@ -31,24 +31,23 @@ class SbertModel(TorchModel, SbertModelTransform):
     def extract_pooled_outputs(self, outputs):
         return outputs['pooler_output']
 
-    def forward(
-        self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        encoder_hidden_states=None,
-        encoder_attention_mask=None,
-        past_key_values=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+    def forward(self,
+                input_ids=None,
+                attention_mask=None,
+                token_type_ids=None,
+                position_ids=None,
+                head_mask=None,
+                inputs_embeds=None,
+                encoder_hidden_states=None,
+                encoder_attention_mask=None,
+                past_key_values=None,
+                use_cache=None,
+                output_attentions=None,
+                output_hidden_states=None,
+                return_dict=None,
+                **kwargs):
         return SbertModelTransform.forward(
             self, input_ids, attention_mask, token_type_ids, position_ids,
             head_mask, inputs_embeds, encoder_hidden_states,
             encoder_attention_mask, past_key_values, use_cache,
-            output_attentions, output_hidden_states, return_dict)
+            output_attentions, output_hidden_states, return_dict, **kwargs)
diff --git a/modelscope/models/nlp/structbert/modeling_sbert.py b/modelscope/models/nlp/structbert/modeling_sbert.py
index 10c0821c..e789037a 100755
--- a/modelscope/models/nlp/structbert/modeling_sbert.py
+++ b/modelscope/models/nlp/structbert/modeling_sbert.py
@@ -870,22 +870,21 @@ class SbertModel(SbertPreTrainedModel):
         output_type=BaseModelOutputWithPoolingAndCrossAttentions,
         config_class=_CONFIG_FOR_DOC,
     )
-    def forward(
-        self,
-        input_ids=None,
-        attention_mask=None,
-        token_type_ids=None,
-        position_ids=None,
-        head_mask=None,
-        inputs_embeds=None,
-        encoder_hidden_states=None,
-        encoder_attention_mask=None,
-        past_key_values=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+    def forward(self,
+                input_ids=None,
+                attention_mask=None,
+                token_type_ids=None,
+                position_ids=None,
+                head_mask=None,
+                inputs_embeds=None,
+                encoder_hidden_states=None,
+                encoder_attention_mask=None,
+                past_key_values=None,
+                use_cache=None,
+                output_attentions=None,
+                output_hidden_states=None,
+                return_dict=None,
+                **kwargs):
         r"""
         encoder_hidden_states  (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`,
             `optional`):