Update handler.py

handler.py

@@ -16,6 +16,7 @@ class EndpointHandler():
         # self.pipeline = pipeline(task= "automatic-speech-recognition", model=path)
         # self.pipeline.model.config.forced_decoder_ids = self.pipeline.tokenizer.get_decoder_prompt_ids(language="Chinese", task="transcribe")
         # self.pipeline.model.generation_config.forced_decoder_ids = self.pipeline.model.config.forced_decoder_ids  # just to be sure!
+        
         peft_model_id = "cathyi/openai-whisper-large-v2-Lora"
         language = "Chinese"
         task = "transcribe"
@@ -23,13 +24,13 @@ class EndpointHandler():
         model = WhisperForConditionalGeneration.from_pretrained(
             peft_config.base_model_name_or_path
         )
-        model = PeftModel.from_pretrained(model, peft_model_id)
+        model = PeftModel.from_pretrained(model, path)
         tokenizer = WhisperTokenizer.from_pretrained(peft_config.base_model_name_or_path, language=language, task=task)
         processor = WhisperProcessor.from_pretrained(peft_config.base_model_name_or_path, language=language, task=task)
         feature_extractor = processor.feature_extractor
         self.forced_decoder_ids = processor.get_decoder_prompt_ids(language=language, task=task)
-        self.pipeline = AutomaticSpeechRecognitionPipeline(model=model, tokenizer=tokenizer, feature_extractor=feature_extractor)
-        # self.pipeline = pipeline(task= "automatic-speech-recognition", model=path, tokenizer=tokenizer, feature_extractor = feature_extractor)
+        # self.pipeline = AutomaticSpeechRecognitionPipeline(model=model, tokenizer=tokenizer, feature_extractor=feature_extractor)
+        self.pipeline = pipeline(task= "automatic-speech-recognition", model=model, tokenizer=tokenizer, feature_extractor = feature_extractor)
     
     def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
         """
@@ -46,8 +47,12 @@ class EndpointHandler():
         print("a1", inputs)
         print("a2", inputs, file=sys.stderr)
         print("a3", inputs, file=sys.stdout)
-        prediction = self.pipeline(inputs, generate_kwargs={"forced_decoder_ids": self.forced_decoder_ids}, max_new_tokens=255)["text"]
-        # prediction = self.pipeline(inputs, return_timestamps=False)
+        try:
+            # prediction = self.pipeline(inputs, generate_kwargs={"forced_decoder_ids": self.forced_decoder_ids}, max_new_tokens=255)["text"]
+            prediction = self.pipeline(inputs, return_timestamps=False)
+        except : 
+            print("error")
+            
         print("b1", prediction)
         print("b2", predcition, file=sys.stderr)
         print("b3", predcition, file=sys.stdout)