Update app.py

app.py

@@ -19,100 +19,133 @@ import json
 import re
 from datetime import datetime
 
-
-
 custom_role_conversions = {"tool-call": "assistant", "tool-response": "user"}
 
 from linkup import LinkupClient
 
 
+def get_image_description(file_name: str, question: str, visual_inspection_tool) -> str:
+    prompt = f"""Write a caption of 5 sentences for this image. Pay special attention to any 
+    details that might be useful for someone answering the following question:
+{question}. But do not try to answer the question directly!
+Do not add any information that is not present in the image."""
+    return visual_inspection_tool(image_path=file_name, question=prompt)
+
+
+def get_document_description(file_path: str, question: str, document_inspection_tool) -> str:
+    prompt = f"""Write a caption of 5 sentences for this document. Pay special attention to any 
+    details that might be useful for someone answering the following question:
+{question}. But do not try to answer the question directly!
+Do not add any information that is not present in the document."""
+    return document_inspection_tool.forward_initial_exam_mode(file_path=file_path, question=prompt)
+
+
+def get_single_file_description(file_path: str, question: str, visual_inspection_tool,
+                                document_inspection_tool):
+    file_extension = file_path.split(".")[-1]
+    if file_extension in ["png", "jpg", "jpeg"]:
+        file_description = f" - Attached image: {file_path}"
+        file_description += (
+            f"\n     -> Image description: "
+            f"{get_image_description(file_path, question, visual_inspection_tool)}"
+        )
+        return file_description
+    elif file_extension in ["pdf", "xls", "xlsx", "docx", "doc", "xml"]:
+        file_description = f" - Attached document: {file_path}"
+        image_path = file_path.split(".")[0] + ".png"
+        if os.path.exists(image_path):
+            description = get_image_description(image_path, question, visual_inspection_tool)
+        else:
+            description = get_document_description(file_path, question, document_inspection_tool)
+        file_description += f"\n     -> File description: {description}"
+        return file_description
+    elif file_extension in ["mp3", "m4a", "wav"]:
+        return f" - Attached audio: {file_path}"
+    else:
+        return f" - Attached file: {file_path}"
 
 
 class BasicAgent:
     def __init__(self):
         """
         Initialize the GAIA dataset agent with SmoLagents.
-        
+
         Args:
             api_key: API key for the LLM provider
             model_name: Name of the LLM model to use
         """
         print("BasicAgent initialized.")
-        
+
         # Initialize the agent
-        self.agent = create_agent()
-    
-    def __call__(self, question: str) -> str:
+        agent_assets = create_agent()
+        self.agent = agent_assets["agent"]
+        self.visual_inspection_tool = agent_assets["visualizer"]
+        self.document_inspection_tool = agent_assets["text_inspection_tool"]
+
+    def __call__(self, question: str, file_name: str = None) -> str:
         """
         Process a question and return an answer.
-        
+
         Args:
             question: The question to answer
-            
+
         Returns:
             The answer to the question
         """
-        print(f"Agent received question (first 50 chars): {question[:50]}...")
-        
+        words = question.split()
+        joined_words = " ".join(words[:20])
+        print(f"Agent received question (first 20 words): {joined_words}...")
+
         # Create a prompt for the agent
-        system_prompt = """
-        You are a knowledgeable assistant answering questions from the GAIA dataset. 
-        Your task is to provide accurate, well-researched answers based on the best available information.
-        Follow these guidelines:
-        1. Use search tools to find relevant information
-        2. Check Wikipedia for background knowledge
-        3. Verify facts when possible
-        4. Be transparent about uncertainty
-        5. Provide clear, concise answers with supporting evidence
-        6. If the question is unclear, ask for clarification
-        7. If information is unavailable, say so rather than making things up
-        
-        Only output the direct answer to the question without any additional information. 
-        For example, if the question is what is the "first name name of the person..." only answer with one word which is "name". 
-        Another example: for question "what is the result of 2+2" only output "4"
-        """
-        
-        full_prompt = f"{system_prompt}\n\nQuestion: {question}\n\nAnswer:"
-        
+
+        full_prompt = """You have one question to answer. It is paramount that you provide a 
+        correct answer.
+        Give it all you can: I know for a fact that you have access to all the relevant tools to 
+        solve it and find the correct answer (the answer does exist). Failure or 'I cannot 
+        answer' or 'None found' will not be tolerated, success will be rewarded.
+        Run verification steps if that's needed, you must make sure you find the correct answer!
+        Here is the task:
+        """ + question
+
+        if file_name:
+            prompt_use_files = ("\n\nTo solve the task above, you will have to use this attached "
+                                "file:")
+            prompt_use_files += get_single_file_description(
+                file_name, question, self.visual_inspection_tool,
+                self.document_inspection_tool
+            )
+            full_prompt += prompt_use_files
+
         try:
             # Run the agent
-            # response = self.agent.run(full_prompt)
-            client = LinkupClient()
-
-            res = client.search(
-              query=full_prompt,
-              depth="deep",
-              output_type="sourcedAnswer",
-              include_images=False,
-            )
-            response = getattr(res, "answer", "No answer provided.")
-            
-            # print(response)
-            
+            response = self.agent.run(full_prompt)
+
             # Clean up the response
             # Remove any system-prompt-like text at the beginning
             cleaned_response = re.sub(r'^.*?Answer:', '', response, flags=re.DOTALL).strip()
-            
+
             if not cleaned_response:
                 cleaned_response = response  # Fallback to original if cleaning removes everything
-            
-            print(f"Agent returning answer (first 50 chars): {cleaned_response[:50]}...")
+            words = cleaned_response.split()
+            joined_words = " ".join(words[:20])
+            print(f"Agent returning answer (first 20 words): {joined_words}...")
             return cleaned_response
         except Exception as e:
             error_msg = f"Error processing question: {str(e)}"
             print(error_msg)
             return error_msg
 
-def run_and_submit_all( profile: gr.OAuthProfile | None):
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
     Fetches all questions, runs the BasicAgent on them, submits all answers,
     and displays the results.
     """
     # --- Determine HF Space Runtime URL and Repo URL ---
-    space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
+    space_id = os.getenv("SPACE_ID")  # Get the SPACE_ID for sending link to the code
 
     if profile:
-        username= f"{profile.username}"
+        username = f"{profile.username}"
         print(f"User logged in: {username}")
     else:
         print("User not logged in.")
@@ -128,7 +161,8 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     except Exception as e:
         print(f"Error instantiating agent: {e}")
         return f"Error initializing agent: {e}", None
-    # In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
+    # In the case of an app running as a hugging Face space, this link points toward your 
+    # codebase ( usefull for others so please keep it public)
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
     print(agent_code)
 
@@ -139,16 +173,16 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         response.raise_for_status()
         questions_data = response.json()
         if not questions_data:
-             print("Fetched questions list is empty.")
-             return "Fetched questions list is empty or invalid format.", None
+            print("Fetched questions list is empty.")
+            return "Fetched questions list is empty or invalid format.", None
         print(f"Fetched {len(questions_data)} questions.")
     except requests.exceptions.RequestException as e:
         print(f"Error fetching questions: {e}")
         return f"Error fetching questions: {e}", None
     except requests.exceptions.JSONDecodeError as e:
-         print(f"Error decoding JSON response from questions endpoint: {e}")
-         print(f"Response text: {response.text[:500]}")
-         return f"Error decoding server response for questions: {e}", None
+        print(f"Error decoding JSON response from questions endpoint: {e}")
+        print(f"Response text: {response.text[:500]}")
+        return f"Error decoding server response for questions: {e}", None
     except Exception as e:
         print(f"An unexpected error occurred fetching questions: {e}")
         return f"An unexpected error occurred fetching questions: {e}", None
@@ -158,27 +192,31 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     answers_payload = []
     print(f"Running agent on {len(questions_data)} questions...")
     for idx, item in enumerate(questions_data):
-        print(f"Question {idx+1}: {item}")
+        print(f"Question {idx + 1}: {item}")
         task_id = item.get("task_id")
         question_text = item.get("question")
         if not task_id or question_text is None:
             print(f"Skipping item with missing task_id or question: {item}")
             continue
         try:
-            submitted_answer = "hello"#agent(question_text)
+            submitted_answer = "hello"  # agent(question_text)
             answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text,
+                                "Submitted Answer": submitted_answer})
         except Exception as e:
-             print(f"Error running agent on task {task_id}: {e}")
-             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+            print(f"Error running agent on task {task_id}: {e}")
+            results_log.append({"Task ID": task_id, "Question": question_text,
+                                "Submitted Answer": f"AGENT ERROR: {e}"})
 
     if not answers_payload:
         print("Agent did not produce any answers to submit.")
         return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
 
-    # 4. Prepare Submission 
-    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
-    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
+    # 4. Prepare Submission
+    submission_data = {"username": username.strip(), "agent_code": agent_code,
+                       "answers": answers_payload}
+    status_update = (f"Agent finished. Submitting {len(answers_payload)} answers for user '"
+                     f"{username}'...")
     print(status_update)
 
     # 5. Submit
@@ -191,7 +229,8 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
             f"Submission Successful!\n"
             f"User: {result_data.get('username')}\n"
             f"Overall Score: {result_data.get('score', 'N/A')}% "
-            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
+            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', 
+                                                                        '?')} correct)\n"
             f"Message: {result_data.get('message', 'No message received.')}"
         )
         print("Submission successful.")
@@ -232,14 +271,21 @@ with gr.Blocks() as demo:
         """
         **Instructions:**
 
-        1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
-        2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
-        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
+        1.  Please clone this space, then modify the code to define your agent's logic, 
+        the tools, the necessary packages, etc ...
+        2.  Log in to your Hugging Face account using the button below. This uses your HF 
+        username for submission.
+        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, 
+        submit answers, and see the score.
 
         ---
         **Disclaimers:**
-        Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
-        This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
+        Once clicking on the "submit button, it can take quite some time ( this is the time for 
+        the agent to go through all the questions).
+        This space provides a basic setup and is intentionally sub-optimal to encourage you to 
+        develop your own, more robust solution. For instance for the delay process of the submit 
+        button, a solution could be to cache the answers and submit in a seperate action or even 
+        to answer the questions in async.
         """
     )
 
@@ -257,10 +303,10 @@ with gr.Blocks() as demo:
     )
 
 if __name__ == "__main__":
-    print("\n" + "-"*30 + " App Starting " + "-"*30)
+    print("\n" + "-" * 30 + " App Starting " + "-" * 30)
     # Check for SPACE_HOST and SPACE_ID at startup for information
     space_host_startup = os.getenv("SPACE_HOST")
-    space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
+    space_id_startup = os.getenv("SPACE_ID")  # Get SPACE_ID at startup
 
     if space_host_startup:
         print(f"✅ SPACE_HOST found: {space_host_startup}")
@@ -268,14 +314,16 @@ if __name__ == "__main__":
     else:
         print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
 
-    if space_id_startup: # Print repo URLs if SPACE_ID is found
+    if space_id_startup:  # Print repo URLs if SPACE_ID is found
         print(f"✅ SPACE_ID found: {space_id_startup}")
         print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
         print(f"   Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
     else:
-        print("ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")
+        print(
+            "ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be "
+            "determined.")
 
-    print("-"*(60 + len(" App Starting ")) + "\n")
+    print("-" * (60 + len(" App Starting ")) + "\n")
 
     print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+    demo.launch(debug=True, share=False)