migrated to apenai

app.py

@@ -15,10 +15,13 @@ from typing import TypedDict, Annotated
 from langgraph.prebuilt import ToolNode
 from langgraph.graph import START, StateGraph, END, Graph
 # from langgraph.prebuilt import tools_condition
-from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
+# from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
 from langchain.agents import initialize_agent, AgentType
 # from langchain_community.llms import HuggingFaceHub
 # from langchain_community.chat_models import ChatHuggingFace
+import openai
+from openai import OpenAI
+# from langchain_openai import OpenAI
 
 ## # Load environment variables from .env file
 # --- Constants ---
@@ -28,6 +31,9 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 # load_dotenv()
 HF_ACCESS_KEY = os.getenv('HF_ACCESS_KEY')
 WEATHER_API_KEY = os.getenv('WEATHER_API_KEY')
+OPENAI_MODEL =  os.getenv('OPENAI_MODEL') #'gpt-3.5-turbo-0613'
+OPENAI_KEY = os.getenv('OPENAI_KEY')
+client = OpenAI(api_key = OPENAI_KEY)
 
 ########## ----- DEFINING TOOLS -----##########
 
@@ -163,6 +169,52 @@ def currency_convert(args: str) -> str:
     except Exception:
         return "error"
 
+# --- TOOL 9: Image Captioning Tool ---
+@tool
+def image_caption(image_url: str) -> str:
+    """Generate a descriptive caption for an image given its URL."""
+    api_url = "https://api-inference.huggingface.co/models/Salesforce/blip-image-captioning-base"
+    headers = {"Authorization": f"Bearer {HF_ACCESS_KEY}"}
+    payload = {"inputs": image_url}
+    try:
+        resp = requests.post(api_url, headers=headers, json=payload, timeout=30)
+        resp.raise_for_status()
+        data = resp.json()
+        return data[0]["generated_text"] if isinstance(data, list) else data.get("generated_text", "no_caption")
+    except Exception:
+        return "error"
+    
+# --- TOOL 10: Optical Character Recognition (OCR) Tool ---
+@tool
+def ocr_image(image_url: str) -> str:
+    """Extract text from an image given its URL."""
+    api_url = "https://api-inference.huggingface.co/models/impira/layoutlm-document-qa"
+    headers = {"Authorization": f"Bearer {HF_ACCESS_KEY}"}
+    payload = {"inputs": {"image": image_url, "question": "What text is in the image?"}}
+    try:
+        resp = requests.post(api_url, headers=headers, json=payload, timeout=30)
+        resp.raise_for_status()
+        data = resp.json()
+        return data.get("answer", "no_text_found")
+    except Exception:
+        return "error"
+    
+# --- TOOL 11: Image Classification Tool ---
+@tool
+def classify_image(image_url: str) -> str:
+    """Classify the main object or scene in an image given its URL."""
+    api_url = "https://api-inference.huggingface.co/models/google/vit-base-patch16-224"
+    headers = {"Authorization": f"Bearer {HF_ACCESS_KEY}"}
+    payload = {"inputs": image_url}
+    try:
+        resp = requests.post(api_url, headers=headers, json=payload, timeout=30)
+        resp.raise_for_status()
+        data = resp.json()
+        return data[0]["label"] if isinstance(data, list) else data.get("label", "no_label")
+    except Exception:
+        return "error"
+
+
 ##-- Tool Discovery ---
 # Use @tool for each function.
 # Use get_all_tools() to auto-discover all decorated tools.
@@ -177,6 +229,9 @@ tools_list = [
     wikipedia_summary,
     dictionary_lookup,
     currency_convert,
+    image_caption,
+    ocr_image,
+    classify_image
 ]
 
 tool_descriptions = "\n".join(f"- {tool.name}: {tool.description}" for tool in tools_list)
@@ -216,22 +271,31 @@ Instructions:
 
 ## --- Initialize Hugging Face Model ---
 # Generate the chat interface, including the tools
+'''
 llm = HuggingFaceEndpoint(
     repo_id="Qwen/Qwen2.5-Coder-32B-Instruct",
     huggingfacehub_api_token=HF_ACCESS_KEY,
     model_kwargs={'prompt': system_prompt}
     # system_prompt=system_prompt,
 )
-
 chat_llm = ChatHuggingFace(llm=llm)
+'''
+llm = OpenAI(
+    openai_api_key=OPENAI_KEY,
+    model_name=OPENAI_MODEL,
+    temperature=0.1
+)
+
 # chat = ChatHuggingFace(llm=llm, verbose=True)
 # tools = [search_tool, fetch_weather]
 # chat_with_tools = chat.bind_tools(tools)
 
 agent = initialize_agent(
     tools=tools_list,
-    llm=chat_llm,
+    llm=llm,
+    # llm=chat_llm,
     agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
+    agent_kwargs={"system_message": system_prompt},
     verbose=True,
     handle_parsing_errors=True
 )
@@ -301,8 +365,9 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
             print(f"Skipping item with missing task_id or question: {item}")
             continue
         try:
-            full_prompt = f"{system_prompt}\n Input Question: {question_text}"
-            submitted_answer = agent(full_prompt)
+            # full_prompt = f"{system_prompt}\n Input Question: {question_text}"
+            # submitted_answer = agent.run(full_prompt)
+            submitted_answer = agent.run(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})
         except Exception as e:

requirements.txt

@@ -6,4 +6,5 @@ langchainhub
 huggingface-hub
 langchain-huggingface
 langchain-community
-transformers
+transformers
+openai