[EncoderDecoderModel] add a add_cross_attention boolean to config (#6377)

* correct encoder decoder model

* Apply suggestions from code review

* apply sylvains suggestions

src/transformers/configuration_encoder_decoder.py

@@ -58,6 +58,7 @@ class EncoderDecoderConfig(PretrainedConfig):
             >>> config_decoder  = model.config.decoder
             >>> # set decoder config to causal lm
             >>> config_decoder.is_decoder = True
+            >>> config_decoder.add_cross_attention = True
 
             >>> # Saving the model, including its configuration
             >>> model.save_pretrained('my-model')
@@ -94,8 +95,9 @@ class EncoderDecoderConfig(PretrainedConfig):
         Returns:
             :class:`EncoderDecoderConfig`: An instance of a configuration object
         """
-        logger.info("Set `config.is_decoder=True` for decoder_config")
+        logger.info("Set `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config")
         decoder_config.is_decoder = True
+        decoder_config.add_cross_attention = True
 
         return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict())
 

src/transformers/configuration_utils.py

@@ -56,6 +56,8 @@ class PretrainedConfig(object):
                 Whether the model is used as an encoder/decoder or not.
             is_decoder (:obj:`bool`, `optional`, defaults to :obj:`False`):
                 Whether the model is used as decoder or not (in which case it's used as an encoder).
+            add_cross_attention (:obj:`bool`, `optional`, defaults to :obj:`False`):
+                Whether cross-attention layers should be added to the model. Note, this option is only relevant for models that can be used as decoder models within the `:class:~transformers.EncoderDecoderModel` class, which consists of all models in ``AUTO_MODELS_FOR_CAUSAL_LM``.
             prune_heads (:obj:`Dict[int, List[int]]`, `optional`, defaults to :obj:`{}`):
                 Pruned heads of the model. The keys are the selected layer indices and the associated values, the list
                 of heads to prune in said layer.
@@ -145,6 +147,7 @@ class PretrainedConfig(object):
         # Is decoder is used in encoder-decoder models to differentiate encoder from decoder
         self.is_encoder_decoder = kwargs.pop("is_encoder_decoder", False)
         self.is_decoder = kwargs.pop("is_decoder", False)
+        self.add_cross_attention = kwargs.pop("add_cross_attention", False)
 
         # Parameters for sequence generation
         self.max_length = kwargs.pop("max_length", 20)

src/transformers/modeling_bert.py

@@ -378,7 +378,9 @@ class BertLayer(nn.Module):
         super().__init__()
         self.attention = BertAttention(config)
         self.is_decoder = config.is_decoder
-        if self.is_decoder:
+        self.add_cross_attention = config.add_cross_attention
+        if self.add_cross_attention:
+            assert self.is_decoder, f"{self} should be used as a decoder model if cross attention is added"
             self.crossattention = BertAttention(config)
         self.intermediate = BertIntermediate(config)
         self.output = BertOutput(config)
@@ -399,6 +401,9 @@ class BertLayer(nn.Module):
         outputs = self_attention_outputs[1:]  # add self attentions if we output attention weights
 
         if self.is_decoder and encoder_hidden_states is not None:
+            assert hasattr(
+                self, "crossattention"
+            ), f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers by setting `config.add_cross_attention=True`"
             cross_attention_outputs = self.crossattention(
                 attention_output,
                 attention_mask,
@@ -695,8 +700,10 @@ class BertModel(BertPreTrainedModel):
     Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
 
     To behave as an decoder the model needs to be initialized with the
-    :obj:`is_decoder` argument of the configuration set to :obj:`True`; an
-    :obj:`encoder_hidden_states` is expected as an input to the forward pass.
+    :obj:`is_decoder` argument of the configuration set to :obj:`True`.
+    To be used in a Seq2Seq model, the model needs to initialized with both :obj:`is_decoder`
+    argument and :obj:`add_cross_attention` set to :obj:`True`; an
+    :obj:`encoder_hidden_states` is then expected as an input to the forward pass.
 
     .. _`Attention is all you need`:
         https://arxiv.org/abs/1706.03762

src/transformers/modeling_encoder_decoder.py

@@ -168,17 +168,18 @@ class EncoderDecoderModel(PreTrainedModel):
                 from .configuration_auto import AutoConfig
 
                 decoder_config = AutoConfig.from_pretrained(decoder_pretrained_model_name_or_path)
-                if decoder_config.is_decoder is False:
+                if decoder_config.is_decoder is False or decoder_config.add_cross_attention is False:
                     logger.info(
                         f"Initializing {decoder_pretrained_model_name_or_path} as a decoder model. Cross attention layers are added to {decoder_pretrained_model_name_or_path} and randomly initialized if {decoder_pretrained_model_name_or_path}'s architecture allows for cross attention layers."
                     )
                     decoder_config.is_decoder = True
+                    decoder_config.add_cross_attention = True
 
                 kwargs_decoder["config"] = decoder_config
 
-            if kwargs_decoder["config"].is_decoder is False:
+            if kwargs_decoder["config"].is_decoder is False or decoder_config.add_cross_attention is False:
                 logger.warning(
-                    f"Decoder model {decoder_pretrained_model_name_or_path} is not initialized as a decoder. In order to initialize {decoder_pretrained_model_name_or_path} as a decoder, make sure that the attribute `is_decoder` of `decoder_config` passed to `.from_encoder_decoder_pretrained(...)` is set to `True` or do not pass a `decoder_config` to `.from_encoder_decoder_pretrained(...)`"
+                    f"Decoder model {decoder_pretrained_model_name_or_path} is not initialized as a decoder. In order to initialize {decoder_pretrained_model_name_or_path} as a decoder, make sure that the attributes `is_decoder` and `add_cross_attention` of `decoder_config` passed to `.from_encoder_decoder_pretrained(...)` are set to `True` or do not pass a `decoder_config` to `.from_encoder_decoder_pretrained(...)`"
                 )
 
             decoder = AutoModelForCausalLM.from_pretrained(decoder_pretrained_model_name_or_path, **kwargs_decoder)