Create model.py

model.py

@@ -0,0 +1,63 @@
+import os
+from PyPDF2 import PdfReader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_community.docstore.in_memory import InMemoryDocstore
+from langchain_community.llms import HuggingFaceHub
+from langchain.chains import RetrievalQA
+from langchain.prompts import PromptTemplate
+import uuid
+import faiss
+
+vectorstore = None
+
+def load_vectorstore(pdf_path):
+    global vectorstore
+
+    reader = PdfReader(pdf_path)
+    text = "".join([page.extract_text() or "" for page in reader.pages])
+    splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+    chunks = splitter.split_text(text)
+
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
+    dim = len(embeddings.embed_query("test"))
+    index = faiss.IndexFlatL2(dim)
+
+    vectorstore = FAISS(
+        embedding_function=embeddings,
+        index=index,
+        docstore=InMemoryDocstore({}),
+        index_to_docstore_id={}
+    )
+    uuids = [str(uuid.uuid4()) for _ in chunks]
+    vectorstore.add_texts(chunks, ids=uuids)
+
+
+def ask_question(query):
+    global vectorstore
+    if not vectorstore:
+        return "Please upload and index a document first."
+
+    llm = HuggingFaceHub(
+        repo_id="mistralai/Mistral-7B-Instruct-v0.1",
+        huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN"),
+        model_kwargs={"temperature": 0.7, "max_length": 512}
+    )
+
+    retriever = vectorstore.as_retriever(search_kwargs={"k": 3})
+    prompt = PromptTemplate(
+        template="Use the context to answer the question:
+Context: {context}
+Question: {question}
+Answer:",
+        input_variables=["context", "question"]
+    )
+
+    chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        retriever=retriever,
+        return_source_documents=False,
+        chain_type_kwargs={"prompt": prompt}
+    )
+    return chain({"query": query})["result"]