test1

modeling_florence2.py

@@ -3067,32 +3067,21 @@ class Florence2ForConditionalGeneration(Florence2PreTrainedModel):
         encoder_outputs=None,
         **kwargs,
     ):
-        # cut decoder_input_ids if past_key_values is used
-        if past_key_values is not None:
-            past_length = past_key_values[0][0].shape[2]
-
-            # Some generation methods already pass only the last input ID
-            if decoder_input_ids.shape[1] > past_length:
-                remove_prefix_length = past_length
-            else:
-                # Default to old behavior: keep only final ID
-                remove_prefix_length = decoder_input_ids.shape[1] - 1
-
-            decoder_input_ids = decoder_input_ids[:, remove_prefix_length:]
-        
-        return {
-            "input_ids": None,  # encoder_outputs is defined. input_ids not needed
-            "encoder_outputs": encoder_outputs,
-            "past_key_values": past_key_values,
-            "decoder_input_ids": decoder_input_ids,
-            "attention_mask": attention_mask,
-            "pixel_values": pixel_values,
-            "decoder_attention_mask": decoder_attention_mask,
-            "head_mask": head_mask,
-            "decoder_head_mask": decoder_head_mask,
-            "cross_attn_head_mask": cross_attn_head_mask,
-            "use_cache": use_cache,  # change this to avoid caching (presumably for debugging)
-        }
+        # Ta funkcja powinna delegować wywołanie do `language_model`,
+        # a nie zwracać własny słownik.
+        # Poprawione wywołanie przekazuje wszystkie argumenty.
+        return self.language_model.prepare_inputs_for_generation(
+            decoder_input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            decoder_attention_mask=decoder_attention_mask,
+            head_mask=head_mask,
+            decoder_head_mask=decoder_head_mask,
+            cross_attn_head_mask=cross_attn_head_mask,
+            use_cache=use_cache,
+            encoder_outputs=encoder_outputs,
+            **kwargs,
+        )
     
     def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
         return self.language_model.shift_tokens_right(labels)