Create app.py

app.py

@@ -0,0 +1,99 @@
+import chainlit as cl
+from langchain.vectorstores import FAISS
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.document_loaders import TextLoader
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.chains import RetrievalQA
+from langchain.prompts import PromptTemplate
+from langchain.llms import HuggingFacePipeline
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+import torch
+import os
+
+# === Step 1: Build or load Vectorstore ===
+
+VECTORSTORE_DIR = "vectorstore"
+DATA_PATH = "data/textile_notes.txt"  # Your textile documents path
+
+def build_vectorstore():
+    loader = TextLoader(DATA_PATH)
+    documents = loader.load()
+    splitter = CharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+    chunks = splitter.split_documents(documents)
+    embeddings = HuggingFaceEmbeddings(model_name="BAAI/bge-small-en-v1.5")
+
+    db = FAISS.from_documents(chunks, embeddings)
+    db.save_local(VECTORSTORE_DIR)
+    return db
+
+def load_vectorstore():
+    embeddings = HuggingFaceEmbeddings(model_name="BAAI/bge-small-en-v1.5")
+    return FAISS.load_local(VECTORSTORE_DIR, embeddings)
+
+# === Step 2: Load LLM and create QA chain ===
+
+def load_qa_chain():
+    # Load local vectorstore if exists else build it
+    if os.path.exists(VECTORSTORE_DIR):
+        vectorstore = load_vectorstore()
+    else:
+        vectorstore = build_vectorstore()
+
+    # Load open-source Mistral 7B Instruct model (small & free-ish)
+    tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
+    model = AutoModelForCausalLM.from_pretrained(
+        "mistralai/Mistral-7B-Instruct-v0.1",
+        torch_dtype=torch.float16,
+        device_map="auto"
+    )
+
+    pipe = pipeline(
+        "text-generation",
+        model=model,
+        tokenizer=tokenizer,
+        max_new_tokens=512,
+        do_sample=True,
+        temperature=0.7,
+    )
+    llm = HuggingFacePipeline(pipeline=pipe)
+
+    # Prompt template for friendly, user-focused answers
+    prompt_template = """
+Answer the question using ONLY the context below.
+Be clear, helpful, and friendly.
+
+Context:
+{context}
+
+Question:
+{question}
+"""
+
+    prompt = PromptTemplate(
+        template=prompt_template,
+        input_variables=["context", "question"]
+    )
+
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=vectorstore.as_retriever(search_kwargs={"k": 4}),
+        chain_type_kwargs={"prompt": prompt},
+        return_source_documents=True,
+    )
+
+    return qa_chain
+
+# === Chainlit event handlers ===
+
+@cl.on_chat_start
+async def on_chat_start():
+    qa = load_qa_chain()
+    cl.user_session.set("qa_chain", qa)
+    await cl.Message("👋 Hi! Ask me anything about textile — I'll answer using our custom documents.").send()
+
+@cl.on_message
+async def on_message(message: cl.Message):
+    qa = cl.user_session.get("qa_chain")
+    answer = qa.run(message.content)
+    await cl.Message(answer).send()