initial commit

.env

@@ -1 +1 @@
-GEMINI_API_KEY = "your_gemini_api_key_here"
+HUGGINGFACEHUB_API_TOKEN = "your_huggingface_api_key_here"

.python-version

@@ -1 +1 @@
-3.10
+3.12

modules/rag_query.py

@@ -1,51 +1,48 @@
+import os
 import chromadb
-from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
 from langchain_core.prompts import ChatPromptTemplate
 from langchain_core.runnables import RunnablePassthrough
 from langchain_huggingface.embeddings import HuggingFaceEmbeddings
 from langchain_chroma import Chroma
-import os
-
-from dotenv import load_dotenv
-
-load_dotenv()  # Load environment variables from .env file
 
 # 1. Initialize RAG Components
-GEMINI_MODEL = "gemini-2.5-flash"
-COLLECTION_NAME = 'video_analysis_data' # Updated collection name
+REPO_ID = "Qwen/Qwen2.5-7B-Instruct" 
+COLLECTION_NAME = 'video_analysis_data'
 DB_PATH = "./chroma_db"
 
 def run_query(user_query):
-    """
-    Executes the RAG pipeline: Retrieve relevant context and generate a general answer.
-    """
-    if not os.path.exists(DB_PATH):
-        print(f"Error: Database path {DB_PATH} not found. Run 'rag_indexer.py' first.")
-        return "Analysis data is not yet indexed. Please index the data first."
+    if not os.getenv("HUGGINGFACEHUB_API_TOKEN"):
+        return "Error: Please set HUGGINGFACEHUB_API_TOKEN in your environment variables."
 
-    # ChromaDB setup
     client = chromadb.PersistentClient(path=DB_PATH)
-    
     embedding_function = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
+    
     vectorstore = Chroma(
         client=client, 
-        collection_name=COLLECTION_NAME, 
+        collection_name=COLLECTION_NAME,
         embedding_function=embedding_function
     )
     
-    # 2. Retrieval (R)
-    retriever = vectorstore.as_retriever(search_kwargs={"k": 5}) # Retrieve top 5 relevant documents
+    retriever = vectorstore.as_retriever(search_kwargs={"k": 5})
 
-    # 3. Generation (G)
-    # The new general-purpose system prompt
-    template = """
-    You are an expert Video Content Analyst. Your task is to describe the scene, objects, and potential context detected in a video based on the provided analysis logs (Context).
-    Use the Context to answer the user's Question.
-    
-    IMPORTANT: Since the analysis only provides object labels and locations (Object Detection) and not specific actions (Activity Recognition), you must infer and describe the *scene* or *context* based on the objects present (e.g., "The presence of people and construction materials suggests activity on a construction site.").
+    # 2. LLM Setup (Qwen via Hugging Face Endpoint)
+    llm_endpoint = HuggingFaceEndpoint(
+        repo_id=REPO_ID,
+        task="text-generation",
+        max_new_tokens=512,
+        repetition_penalty=1.1,
+        huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
+    )
     
-    Always include the video timestamp(s) where the relevant context was found.
+    # Wrap it in ChatHuggingFace to handle the prompt templates correctly
+    llm = ChatHuggingFace(llm=llm_endpoint)
 
+    template = """
+    You are an expert Video Content Analyst. Use the Context to answer the Question.
+    If you don't know the answer, say you don't know. 
+    Infer activity based on detected objects (e.g., people + skateboards = skateboarding).
+    
     Context:
     {context}
 
@@ -53,19 +50,14 @@ def run_query(user_query):
     """
     prompt = ChatPromptTemplate.from_template(template)
 
-    llm = ChatGoogleGenerativeAI(model=GEMINI_MODEL)
-    
-    # The RAG Chain logic
     rag_chain = (
         {"context": retriever, "question": RunnablePassthrough()}
         | prompt
         | llm
     )
     
-    # 4. Execute the chain
-    print(f"Executing RAG query for: '{user_query}'...")
     response = rag_chain.invoke(user_query)
-    
+    # Hugging Face responses sometimes need a little cleaning depending on the version
     return response.content
 
 # Example Usage:

uv.lock

@@ -2211,6 +2211,23 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a4/7d/f1c30a92854540bf789e9cd5dde7ef49bbe63f855b85a2e6b3db8135c591/opencv_python-4.11.0.86-cp37-abi3-win_amd64.whl", hash = "sha256:085ad9b77c18853ea66283e98affefe2de8cc4c1f43eda4c100cf9b2721142ec", size = 39488044, upload-time = "2025-01-16T13:52:21.928Z" },
 ]
 
+[[package]]
+name = "opencv-python-headless"
+version = "4.11.0.86"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/36/2f/5b2b3ba52c864848885ba988f24b7f105052f68da9ab0e693cc7c25b0b30/opencv-python-headless-4.11.0.86.tar.gz", hash = "sha256:996eb282ca4b43ec6a3972414de0e2331f5d9cda2b41091a49739c19fb843798", size = 95177929, upload-time = "2025-01-16T13:53:40.22Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dc/53/2c50afa0b1e05ecdb4603818e85f7d174e683d874ef63a6abe3ac92220c8/opencv_python_headless-4.11.0.86-cp37-abi3-macosx_13_0_arm64.whl", hash = "sha256:48128188ade4a7e517237c8e1e11a9cdf5c282761473383e77beb875bb1e61ca", size = 37326460, upload-time = "2025-01-16T13:52:57.015Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/43/68555327df94bb9b59a1fd645f63fafb0762515344d2046698762fc19d58/opencv_python_headless-4.11.0.86-cp37-abi3-macosx_13_0_x86_64.whl", hash = "sha256:a66c1b286a9de872c343ee7c3553b084244299714ebb50fbdcd76f07ebbe6c81", size = 56723330, upload-time = "2025-01-16T13:55:45.731Z" },
+    { url = "https://files.pythonhosted.org/packages/45/be/1438ce43ebe65317344a87e4b150865c5585f4c0db880a34cdae5ac46881/opencv_python_headless-4.11.0.86-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6efabcaa9df731f29e5ea9051776715b1bdd1845d7c9530065c7951d2a2899eb", size = 29487060, upload-time = "2025-01-16T13:51:59.625Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/5c/c139a7876099916879609372bfa513b7f1257f7f1a908b0bdc1c2328241b/opencv_python_headless-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0e0a27c19dd1f40ddff94976cfe43066fbbe9dfbb2ec1907d66c19caef42a57b", size = 49969856, upload-time = "2025-01-16T13:53:29.654Z" },
+    { url = "https://files.pythonhosted.org/packages/95/dd/ed1191c9dc91abcc9f752b499b7928aacabf10567bb2c2535944d848af18/opencv_python_headless-4.11.0.86-cp37-abi3-win32.whl", hash = "sha256:f447d8acbb0b6f2808da71fddd29c1cdd448d2bc98f72d9bb78a7a898fc9621b", size = 29324425, upload-time = "2025-01-16T13:52:49.048Z" },
+    { url = "https://files.pythonhosted.org/packages/86/8a/69176a64335aed183529207ba8bc3d329c2999d852b4f3818027203f50e6/opencv_python_headless-4.11.0.86-cp37-abi3-win_amd64.whl", hash = "sha256:6c304df9caa7a6a5710b91709dd4786bf20a74d57672b3c31f7033cc638174ca", size = 39402386, upload-time = "2025-01-16T13:52:56.418Z" },
+]
+
 [[package]]
 name = "opentelemetry-api"
 version = "1.39.1"
@@ -3132,7 +3149,7 @@ dependencies = [
     { name = "langchain-google-genai" },
     { name = "langchain-huggingface" },
     { name = "numpy" },
-    { name = "opencv-python" },
+    { name = "opencv-python-headless" },
     { name = "python-dotenv" },
     { name = "sentence-transformers" },
     { name = "streamlit" },
@@ -3151,8 +3168,8 @@ requires-dist = [
     { name = "langchain-core", specifier = ">=1.1.3" },
     { name = "langchain-google-genai", specifier = ">=4.0.0" },
     { name = "langchain-huggingface", specifier = ">=1.1.0" },
-    { name = "numpy", specifier = ">=2.3.5" },
-    { name = "opencv-python", specifier = ">=4.11.0.86" },
+    { name = "numpy", specifier = "==2.3.5" },
+    { name = "opencv-python-headless", specifier = ">=4.10.0.84" },
     { name = "python-dotenv", specifier = ">=1.2.1" },
     { name = "sentence-transformers", specifier = ">=5.2.0" },
     { name = "streamlit", specifier = ">=1.52.1" },