Update app.py

app.py

@@ -121,64 +121,7 @@ class SearchEngine:
         return results
 
 
-class Chatbot:
-    def __init__(self, model_name="EleutherAI/gpt-neo-125M"):
-        """
-        Initializes the chatbot with GPT-Neo.
-        """
-        self.tokenizer = AutoTokenizer.from_pretrained(model_name)
-        self.model = AutoModelForCausalLM.from_pretrained(model_name)
-        
-        # Set pad_token to eos_token if not already defined
-        if self.tokenizer.pad_token is None:
-            self.tokenizer.pad_token = self.tokenizer.eos_token
 
-    def generate_response(self, prompt, max_length=100):
-        """
-        Generates a response to a user query using GPT-Neo.
-        """
-        # Tokenize the input prompt
-        inputs = self.tokenizer(prompt, return_tensors="pt", padding=True, truncation=True)
-        
-        # Generate the response
-        outputs = self.model.generate(
-            inputs.input_ids,
-            attention_mask=inputs.attention_mask,  # Pass the attention mask
-            max_length=max_length,
-            num_return_sequences=1,
-            pad_token_id=self.tokenizer.pad_token_id,  # Use the defined pad_token_id
-        )
-        
-        # Decode the generated response
-        response = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
-        return response
-
-    def handle_request(self, prompt):
-        """
-        Handles user requests by determining the intent and delegating to the appropriate function.
-        """
-        # Check if the user wants to search for something
-        if "search" in prompt.lower():
-            query = prompt.lower().replace("search", "").strip()
-            results = search_engine.search(query)
-            return {"type": "search", "results": results}
-
-        # Check if the user wants a summary
-        elif "summarize" in prompt.lower() or "summary" in prompt.lower():
-            text = prompt.lower().replace("summarize", "").replace("summary", "").strip()
-            summary = summarizer.summarize(text)
-            return {"type": "summary", "summary": summary}
-
-        # Check if the user wants to extract topics
-        elif "topics" in prompt.lower() or "topic" in prompt.lower():
-            text = prompt.lower().replace("topics", "").replace("topic", "").strip()
-            topics = topic_extractor.extract_topics([text])
-            return {"type": "topics", "topics": topics.to_dict()}
-
-        # Default to generating a conversational response
-        else:
-            response = self.generate_response(prompt)
-            return {"type": "chat", "response": response}
 
 
 # Initialize models