Update app.py

app.py

@@ -10,7 +10,7 @@ import pandas as pd
 import nltk
 from nltk.sentiment import SentimentIntensityAnalyzer
 from textblob import TextBlob
-from transformers import AutoModelForSequenceClassification, AutoTokenizer, AutoModelForCausalLM
+from transformers import AutoModelForSequenceClassification, AutoTokenizer, AutoModelForCausalLM, PretrainedConfig
 
 warnings.filterwarnings('ignore', category=FutureWarning)
 
@@ -111,9 +111,10 @@ class LLMResponder:
             self.llm = Llama(model_path="path/to/your/gguf/file.gguf", n_ctx=1024)
             self.backend = "llama_cpp"
         else:
-            # Use trust_remote_code=True to load the model despite missing config keys.
-            self.llm_tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-            self.llm_model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True)
+            # Create a dummy config to bypass the missing model_type in config.json.
+            dummy_config = PretrainedConfig.from_dict({"model_type": "llama"})
+            self.llm_tokenizer = AutoTokenizer.from_pretrained(model_name, config=dummy_config, trust_remote_code=True)
+            self.llm_model = AutoModelForCausalLM.from_pretrained(model_name, config=dummy_config, trust_remote_code=True)
             self.backend = "transformers"
 
     def generate_response(self, prompt):
@@ -190,5 +191,4 @@ iface = gr.Interface(
 )
 
 if __name__ == "__main__":
-    # In Hugging Face Spaces, remove share=True.
     iface.launch()