Make model compatible with Transformers v4.49.0 (#40)

- Make model compatible with Transformers v4.49.0 (ae440694f5af374760e184d6c45b5a8ec4f67301)


Co-authored-by: Harry Mellor <[email protected]>

modeling_minicpmo.py

@@ -42,7 +42,7 @@ from transformers import AutoProcessor
 from transformers import BertTokenizerFast
 from transformers import LlamaConfig
 from transformers import LlamaModel
-from transformers import LogitsWarper
+from transformers import LogitsProcessor
 from transformers import PreTrainedModel
 from transformers import Qwen2ForCausalLM
 from transformers import Qwen2PreTrainedModel
@@ -2919,7 +2919,7 @@ class ConditionalChatTTS(PreTrainedModel):
         force_no_stop=False,
         min_new_token=10,
         max_new_token=50,
-        logits_warpers: List[LogitsWarper] = [],
+        logits_warpers: List[LogitsProcessor] = [],
         logits_processors: List[CustomRepetitionPenaltyLogitsProcessorRepeat] = [],
         show_tqdm=False,
     ):
@@ -2937,7 +2937,7 @@ class ConditionalChatTTS(PreTrainedModel):
             eos_token (Union[int, torch.Tensor]): End of sequence token.
             streaming_tts_text_mask (Optional[torch.Tensor], optional): Mask for streaming TTS text. Defaults to None.
             max_new_token (int, optional): Maximum number of new tokens to generate. Defaults to 50.
-            logits_warpers (List[LogitsWarper], optional): List of logits warpers. Defaults to [].
+            logits_warpers (List[LogitsProcessor], optional): List of logits processors. Defaults to [].
             logits_processors (List[CustomRepetitionPenaltyLogitsProcessorRepeat], optional): List of logits processors. Defaults to [].
             show_tqdm (bool, optional): Whether to show progress bar. Defaults to True.