Update app.py

app.py

@@ -15,7 +15,6 @@ from langchain_community.document_loaders import WikipediaLoader
 from langchain_community.utilities import WikipediaAPIWrapper
 from langchain_community.document_loaders import ArxivLoader
 
-# (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
@@ -26,8 +25,7 @@ import io
 import contextlib
 import traceback
 from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
-from smolagents import Tool, CodeAgent, DuckDuckGoSearchTool, FinalAnswerTool, HfApiModel
-
+# from smolagents import Tool, CodeAgent, DuckDuckGoSearchTool, FinalAnswerTool, HfApiModel # These are already imported above
 
 class CodeLlamaTool(Tool):
     name = "code_llama_tool"
@@ -62,6 +60,7 @@ class CodeLlamaTool(Tool):
         )
 
     def forward(self, question: str) -> str:
+        # Corrected: Use self.prompt and then pass it to the pipeline
         self.prompt = f"""You are an AI that uses Python code to answer questions.
 Question: {question}
 Instructions:
@@ -77,7 +76,7 @@ print(5 * math.sqrt(36))
 <final>30.0</final>
 Answer:"""
 
-        response = self.pipeline(prompt)[0]["generated_text"]
+        response = self.pipeline(self.prompt)[0]["generated_text"] # Pass self.prompt
         return self.parse_and_execute(response)
 
     def parse_and_execute(self, response: str) -> str:
@@ -107,11 +106,6 @@ Answer:"""
         except Exception:
             return f"Error executing code:\n{traceback.format_exc()}"
 
-
-
-#from smolagents import Tool
-#from langchain_community.document_loaders import WikipediaLoader
-
 class WikiSearchTool(Tool):
     name = "wiki_search"
     description = "Search Wikipedia for a query and return up to 2 results."
@@ -131,9 +125,6 @@ class WikiSearchTool(Tool):
         )
         return formatted_search_docs
 
-    
-
-
 class StringReverseTool(Tool):
     name = "reverse_message"
     description = "When you received a strange text, try to reverse it and perform action described in reversed message."
@@ -172,10 +163,10 @@ class KeywordsExtractorTool(Tool):
             filtered_words = []
             for w in all_words:
                 if w not in conjunctions:
-                    filtered_words.push(w)
+                    filtered_words.append(w) # Corrected: Use append instead of push
             word_counts = Counter(filtered_words)
             k = 5
-            return heapq.nlargest(k, word_counts.items(), key=lambda x: x[1])
+            return str(heapq.nlargest(k, word_counts.items(), key=lambda x: x[1])) # Corrected: Convert list of tuples to string
         except Exception as e:
             return f"Error during extracting most common words: {e}"
 
@@ -225,8 +216,6 @@ def parse_excel_to_json(task_id: str) -> dict:
             "status": f"Error in parsing Excel file: {str(e)}"
         }
 
-
-
 class VideoTranscriptionTool(Tool):
     """Fetch transcripts from YouTube videos"""
     name = "transcript_video"
@@ -239,14 +228,17 @@ class VideoTranscriptionTool(Tool):
 
     def forward(self, url: str, include_timestamps: bool = False) -> str:
 
-        if "youtube.com/watch" in url:
+        # Corrected: Handle various YouTube URL formats
+        video_id = None
+        if "youtube.com/watch?v=" in url:
             video_id = url.split("v=")[1].split("&")[0]
         elif "youtu.be/" in url:
             video_id = url.split("youtu.be/")[1].split("?")[0]
-        elif len(url.strip()) == 11:  # Direct ID
+        elif len(url.strip()) == 11 and not ("http://" in url or "https://" in url):  # Direct ID
             video_id = url.strip()
-        else:
-            return f"YouTube URL or ID: {url} is invalid!"
+        
+        if not video_id:
+            return f"YouTube URL or ID: {url} is invalid or not supported!"
 
         try:
             transcription = YouTubeTranscriptApi.get_transcript(video_id)
@@ -265,8 +257,8 @@ class VideoTranscriptionTool(Tool):
 
 class BasicAgent:
     def __init__(self):
-        token = os.environ.get("HF_API_TOKEN")
-  
+        token = os.environ.get("HF_TOKEN") # Corrected: Use HF_TOKEN
+
         # Initialize tokenizer
         self.model_id = "codellama/CodeLlama-7b-Instruct-hf"
         self.tokenizer = AutoTokenizer.from_pretrained(self.model_id, token=token)
@@ -288,7 +280,7 @@ class BasicAgent:
         final_answer_tool = FinalAnswerTool()
         video_transcription_tool = VideoTranscriptionTool()
 
-        # ✅ New Llama Tool
+        # New Llama Tool
         code_llama_tool = CodeLlamaTool()
 
         self.system_prompt = f"""
@@ -304,7 +296,7 @@ Answer should be a short string, number, or comma-separated list. Keep it brief.
                 keywords_extract_tool, speech_to_text_tool,
                 visit_webpage_tool, final_answer_tool,
                 parse_excel_to_json, video_transcription_tool,
-                code_llama_tool  # 🔧 Add here
+                code_llama_tool
             ],
             add_base_tools=True
         )