Upload folder using huggingface_hub

qwen_app.py

@@ -0,0 +1,153 @@
+import os
+from typing import List, Any
+from chainlit.types import AskFileResponse
+import tempfile
+import shutil
+
+# Text processing
+from langchain.text_splitter import CharacterTextSplitter
+from langchain_community.document_loaders import TextLoader, PyPDFLoader
+from langchain.docstore.document import Document
+
+# Prompt templates
+from langchain.prompts import ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate
+
+# Embeddings + VectorDB
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+
+# OpenRouter (Qwen via OpenAI-compatible API)
+from langchain_openai import ChatOpenAI
+
+# UI framework
+import chainlit as cl
+
+
+# -------------------------
+# API Key Setup
+# -------------------------
+# Make sure you export in Colab / Terminal before running:
+#   os.environ["OPENROUTER_API_KEY"] = "your_api_key_here"
+#   os.environ["OPENAI_API_BASE"] = "https://openrouter.ai/api/v1"
+
+
+# -------------------------
+# File processing
+# -------------------------
+text_splitter = CharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+
+def process_file(file: AskFileResponse) -> List[Document]:
+    """Load and split PDF or TXT into LangChain Documents."""
+    suffix = f".{file.name.split('.')[-1]}"
+    with tempfile.NamedTemporaryFile(delete=False, suffix=suffix) as temp_file:
+        shutil.copyfile(file.path, temp_file.name)
+
+        if file.name.lower().endswith(".pdf"):
+            loader = PyPDFLoader(temp_file.name)
+        else:
+            loader = TextLoader(temp_file.name)
+
+        try:
+            docs = loader.load()
+            texts = text_splitter.split_documents(docs)
+            return texts
+        finally:
+            try:
+                os.unlink(temp_file.name)
+            except Exception as e:
+                print(f"Cleanup error: {e}")
+
+
+# -------------------------
+# Retrieval QA Pipeline
+# -------------------------
+class RetrievalAugmentedQAPipeline:
+    def __init__(self, llm: Any, vectorstore: FAISS) -> None:
+        self.llm = llm
+        self.vectorstore = vectorstore
+
+        # Prompt definition
+        system_template = (
+            "You are a helpful assistant. "
+            "Use the following context to answer a user's question. "
+            "If the context does not contain the answer, reply with 'I don't know'."
+        )
+        self.prompt = ChatPromptTemplate.from_messages([
+            SystemMessagePromptTemplate.from_template(system_template),
+            HumanMessagePromptTemplate.from_template("Context:\n{context}\n\nQuestion:\n{question}")
+        ])
+
+    async def arun_pipeline(self, user_query: str):
+        # Retrieve documents
+        docs = self.vectorstore.similarity_search(user_query, k=4)
+        context_text = "\n".join([doc.page_content for doc in docs])
+
+        # Format the prompt
+        messages = self.prompt.format_messages(context=context_text, question=user_query)
+
+        # Stream response from Qwen
+        async def generate_response():
+            async for chunk in self.llm.astream(messages):
+                yield chunk.content if chunk.content else ""
+
+        return {"response": generate_response(), "context": docs}
+
+
+# -------------------------
+# Chainlit Handlers
+# -------------------------
+@cl.on_chat_start
+async def on_chat_start():
+    files = None
+
+    # Wait for user file
+    while files is None:
+        files = await cl.AskFileMessage(
+            content="Please upload a Text or PDF file to begin!",
+            accept=["text/plain", "application/pdf"],
+            max_size_mb=5,
+            timeout=180,
+        ).send()
+
+    file = files[0]
+    msg = cl.Message(content=f"Processing `{file.name}`...")
+    await msg.send()
+
+    # Load & process file
+    texts = process_file(file)
+    print(f"Processing {len(texts)} chunks")
+
+    # Create embeddings + vectorstore
+    embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
+    vectorstore = FAISS.from_documents(texts, embeddings)
+
+    # Initialize Qwen via OpenRouter
+    chat_llm = ChatOpenAI(
+        model="qwen/qwen2.5-vl-72b-instruct",  # ✅ you can swap with qwen-3 when available
+        streaming=True,
+        temperature=0,
+        max_tokens=1024,
+        openai_api_base=os.environ.get("OPENAI_API_BASE", "https://openrouter.ai/api/v1"),
+        openai_api_key= "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"
+    )
+
+    # Create retrieval pipeline
+    retrieval_qa = RetrievalAugmentedQAPipeline(llm=chat_llm, vectorstore=vectorstore)
+
+    msg.content = f"Processing `{file.name}` done ✅. You can now ask questions!"
+    await msg.update()
+
+    cl.user_session.set("chain", retrieval_qa)
+
+
+@cl.on_message
+async def main(message: cl.Message):
+    chain = cl.user_session.get("chain")
+
+    msg = cl.Message(content="")
+    result = await chain.arun_pipeline(message.content)
+
+    async for stream_resp in result["response"]:
+        await msg.stream_token(stream_resp)
+
+    await msg.send()