Update agent.py

agent.py

@@ -8,11 +8,23 @@ class GaiaAgent:
     def __init__(self):
         self.client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
         self.instructions = (
-            "You are an expert assistant solving GAIA benchmark questions. "
-            "You analyze file contents (like CSV), reason step-by-step, and respond with a single factual answer."
+            "You are solving GAIA benchmark questions. "
+            "If given a table, analyze it and extract relevant facts to answer accurately. "
+            "Provide only the final answer. Do not explain."
         )
         self.api_url = "https://agents-course-unit4-scoring.hf.space"
 
+    def summarize_csv(self, csv_text: str) -> str:
+        try:
+            df = pd.read_csv(io.StringIO(csv_text))
+            summary = f"Rows: {len(df)}, Columns: {len(df.columns)}\n"
+            summary += f"Columns: {', '.join(df.columns[:10])}\n"
+            sample_row = df.iloc[0].to_dict()
+            summary += f"Sample row: {sample_row}"
+            return summary
+        except Exception as e:
+            return f"[Failed to parse CSV: {e}]"
+
     def fetch_file_context(self, task_id: str) -> str:
         try:
             url = f"{self.api_url}/files/{task_id}"
@@ -20,36 +32,27 @@ class GaiaAgent:
             response.raise_for_status()
             content_type = response.headers.get("Content-Type", "")
 
-            if "text/csv" in content_type or url.endswith(".csv"):
-                df = pd.read_csv(io.StringIO(response.text))
-                if df.shape[1] <= 15 and df.shape[0] <= 30:
-                    return f"CSV table preview:\n{df.to_markdown(index=False)}"
-                else:
-                    return f"CSV summary: {df.shape[0]} rows, {df.shape[1]} columns.\nColumns: {', '.join(df.columns[:10])}"
-
-            elif "application/json" in content_type:
-                return f"JSON content:\n{response.text[:2000]}"
-
-            elif "application/pdf" in content_type:
-                return "[PDF detected. You may need to request OCR summary.]"
-
+            if "csv" in content_type or url.endswith(".csv"):
+                return self.summarize_csv(response.text)
+            elif "json" in content_type:
+                return f"JSON Sample: {response.text[:1000]}"
             elif "text/plain" in content_type:
-                return f"File preview:\n{response.text[:2000]}"
-
+                return f"Text Sample: {response.text[:1000]}"
+            elif "pdf" in content_type:
+                return "[PDF detected. OCR not supported in this version.]"
             else:
                 return f"[Unsupported file type: {content_type}]"
 
         except Exception as e:
-            return f"[Error downloading or processing file: {e}]"
+            return f"[File fetch error: {e}]"
 
     def __call__(self, question: str, task_id: str = None) -> str:
-        file_context = ""
+        context = ""
         if task_id:
-            file_context = self.fetch_file_context(task_id)
-            if file_context:
-                file_context = f"FILE CONTEXT:\n{file_context}\n"
+            context = self.fetch_file_context(task_id)
+            context = f"FILE DATA:\n{context}\n"
 
-        prompt = f"{self.instructions}\n\n{file_context}QUESTION:\n{question}\nANSWER:"
+        prompt = f"{self.instructions}\n\n{context}QUESTION: {question}\nANSWER:"
 
         response = self.client.chat.completions.create(
             model="gpt-4-turbo",