Update app.py

app.py

@@ -25,7 +25,11 @@ from transformers import BlipProcessor, BlipForConditionalGeneration
 
 import re
 from collections import defaultdict
-
+from pytube import YouTube
+import wikipediaapi
+from langchain.agents import initialize_agent, Tool
+from langchain.llms import HuggingFaceHub
+from langchain.tools import PythonREPL
 
 
 # (Keep Constants as is)
@@ -34,137 +38,98 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
 # --- Basic Agent Definition ---
 # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-# Summarization pipeline (load once)
-summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
-
-
- 
 
-
-class SmartAgentV2:
+# --- Agent Tools ---
+def wikipedia_lookup(query):
+    wiki_wiki = wikipediaapi.Wikipedia('en')
+    page = wiki_wiki.page(query)
+    if not page.exists():
+        return f"Wikipedia page for '{query}' not found."
+    return page.summary[:1024]
+
+wiki_tool = Tool(
+    name="WikipediaTool",
+    func=wikipedia_lookup,
+    description="Use for looking up facts or summaries from English Wikipedia."
+)
+
+def get_youtube_transcript(url):
+    try:
+        yt = YouTube(url)
+        caption = yt.captions.get_by_language_code('en')
+        return caption.generate_srt_captions()[:2048]
+    except Exception as e:
+        return f"Failed to retrieve transcript: {str(e)}"
+
+youtube_tool = Tool(
+    name="YouTubeTranscriptTool",
+    func=get_youtube_transcript,
+    description="Use to retrieve English captions from a YouTube video URL."
+)
+
+def transcribe_audio(file_path):
+    model = whisper.load_model("base")
+    result = model.transcribe(file_path)
+    return result['text'][:2048]
+
+audio_tool = Tool(
+    name="AudioTranscriber",
+    func=transcribe_audio,
+    description="Transcribes short English audio files (MP3/WAV)."
+)
+
+def extract_food_sales(filepath):
+    wb = openpyxl.load_workbook(filepath)
+    sheet = wb.active
+    total = 0
+    for row in sheet.iter_rows(min_row=2, values_only=True):
+        item, category, sales = row
+        if category.lower() == 'food':
+            total += float(sales)
+    return f"Total food sales: ${total:.2f}"
+
+excel_tool = Tool(
+    name="ExcelFoodSales",
+    func=extract_food_sales,
+    description="Use to calculate total food sales from an Excel file with columns: item, category, sales."
+)
+
+def describe_image(image_path):
+    processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
+    model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
+    raw_image = Image.open(image_path).convert('RGB')
+    inputs = processor(raw_image, return_tensors="pt")
+    out = model.generate(**inputs)
+    caption = processor.decode(out[0], skip_special_tokens=True)
+    return caption
+
+image_tool = Tool(
+    name="ImageDescriber",
+    func=describe_image,
+    description="Use to describe an image (e.g., chessboard layout or other visual input)."
+)
+
+repl_tool = PythonREPL()
+
+llm = HuggingFaceHub(repo_id="google/flan-t5-xl", model_kwargs={"temperature": 0.2, "max_length": 1024})
+
+tools = [wiki_tool, youtube_tool, audio_tool, excel_tool, image_tool, repl_tool]
+agent_instance = initialize_agent(tools, llm, agent="zero-shot-react-description", verbose=True)
+
+# --- Enhanced Agent ---
+class BasicAgent:
     def __init__(self):
-        self.qa_model = pipeline("text2text-generation", model="google/flan-t5-xl")
-        self.summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
-        self.whisper_model = whisper.load_model("base")
-
-    def search_web(self, query):
-        with DDGS() as ddgs:
-            results = ddgs.text(query, max_results=3)
-            for r in results:
-                if "href" in r:
-                    return r["href"]
-        return "No results found."
-
-    def summarize_url(self, url):
+        print("Advanced GAIA Agent initialized.")
+    def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
         try:
-            article = Article(url)
-            article.download()
-            article.parse()
-            text = article.text
-            if not text.strip():
-                return "No content found."
-            summary = self.summarizer(text, max_length=150, min_length=40, do_sample=False)
-            return summary[0]['summary_text'].strip()
+            result = agent_instance.run(question)
+            print(f"Agent response: {result[:100]}")
+            return result
         except Exception as e:
-            logging.error(f"Summarization error: {e}")
-            return "Error summarizing."
-
-    def generate_citation(self, url):
-        citation_id = hashlib.md5(url.encode()).hexdigest()[:6]
-        year = datetime.datetime.now().year
-        return f"@article{{cite{citation_id}, title={{Generated Citation}}, author={{Unknown}}, journal={{Online}}, year={{ {year} }}, url={{ {url} }} }}"
-
-    def transcribe_audio(self, filepath):
-        result = self.whisper_model.transcribe(filepath)
-        return result["text"]
-
-    def extract_ingredients(self, transcript):
-        ingredients = re.findall(r"(?:\ba|\ban|\bthe)?\s*([a-zA-Z\s]+?)\s*(?:\bof\b|\bcups?\b|\btablespoons?\b|\bteaspoons?\b|\bpinch\b)?", transcript)
-        ingredients = [i.strip().lower() for i in ingredients if len(i.strip()) > 2]
-        return ", ".join(sorted(set(ingredients)))
-
-    def extract_page_numbers(self, transcript):
-        numbers = re.findall(r"\b\d+\b", transcript)
-        return ", ".join(sorted(set(numbers), key=int))
-
-    def sum_food_sales(self, filepath):
-        df = pd.read_excel(filepath)
-        food_df = df[df["Category"].str.lower() == "food"]
-        total = food_df["Sales"].sum()
-        return f"${total:.2f}"
-
-    def answer_fact(self, question):
-        return self.qa_model(question, max_length=100)[0]["generated_text"].strip()
-
-    def reverse_text_puzzle(self, line):
-        try:
-            return ''.join(reversed(line.strip()))
-        except:
-            return "Could not reverse text."
-
-    def non_commutative_subset(self):
-        return "a, b, c"
-
-    def true_vegetables(self):
-        vegetables = [
-            "broccoli", "celery", "green beans", "lettuce", "sweet potatoes", "zucchini"
-        ]
-        return ", ".join(sorted(vegetables))
-
-    def get_wikipedia_answer(self, question):
-        try:
-            search_url = self.search_web(question)
-            response = requests.get(search_url, timeout=10)
-            soup = BeautifulSoup(response.text, 'html.parser')
-            paragraphs = soup.find_all('p')
-            full_text = ' '.join(p.text for p in paragraphs[:5])
-            answer = self.qa_model(question + "\n" + full_text, max_length=100)[0]['generated_text']
-            return answer.strip()
-        except Exception as e:
-            logging.error(f"Wikipedia fallback failed: {e}")
-            return "Could not find answer from Wikipedia."
-
-    def __call__(self, question: str, file=None):
-        q = question.lower().strip()
-        try:
-            if any(word in q for word in ["image", "chess", "diagram"]):
-                return "I'm a text-only agent and cannot interpret images."
-            if any(word in q for word in ["youtube", "video"]):
-                return "I'm unable to access or analyze video/audio from YouTube."
-            if 'etirw ,ecnetnes' in q:
-                return self.reverse_text_puzzle(question)
-            if "counter-examples" in q and "commutative" in q:
-                return self.non_commutative_subset()
-            if "vegetables" in q and "botany" in q:
-                return self.true_vegetables()
-            if file:
-                if filepath := getattr(file, "name", None):
-                    if filepath.endswith(".mp3"):
-                        transcript = self.transcribe_audio(filepath)
-                        if "ingredient" in q:
-                            return self.extract_ingredients(transcript)
-                        if "page" in q:
-                            return self.extract_page_numbers(transcript)
-                        return transcript
-                    elif filepath.endswith(".xlsx") or filepath.endswith(".xls"):
-                        return self.sum_food_sales(filepath)
-            if q.startswith("summarize:"):
-                url = question.split(":", 1)[1].strip()
-                return self.summarize_url(url)
-            elif q.startswith("generate citation:") or q.startswith("cite:"):
-                url = question.split(":", 1)[1].strip()
-                return self.generate_citation(url)
-            elif q.startswith("search:"):
-                query = question.split(":", 1)[1].strip()
-                return self.search_web(query)
-            elif "wikipedia" in q:
-                return self.get_wikipedia_answer(question)
-            else:
-                return self.answer_fact(question)
-        except Exception as e:
-            logging.error(f"Error: {e}")
-            return "An error occurred processing the question."
-
+            error_message = f"ERROR: {e}"
+            print(error_message)
+            return error_message