Initial commit

app.py

@@ -0,0 +1,24 @@
+import gradio as gr
+from indexer import index_text, answer_query
+
+# Gradio interface function to handle the RAG system
+def rag_system(input_text, query):
+    # Index the input text
+    vectorstore = index_text(input_text)
+    
+    # Answer the query based on the indexed text
+    answer = answer_query(query, vectorstore)
+    
+    return answer
+
+# Build the Gradio interface
+iface = gr.Interface(
+    fn=rag_system, 
+    inputs=["text", "text"], 
+    outputs="text", 
+    title="RAG QA System",
+    description="Enter a text and ask questions based on the input text."
+)
+
+# Launch the app
+iface.launch()

indexer.py

@@ -0,0 +1,32 @@
+from transformers import pipeline
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_core.documents import Document
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+
+# Function to index the input text
+def index_text(input_text):
+    # Split the input text into chunks
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+    splits = text_splitter.split_text(input_text)  
+    documents = [Document(page_content=chunk) for chunk in splits]
+
+    # Embed the input text chunks
+    embeddings = HuggingFaceEmbeddings()
+    
+    vectorstore = FAISS.from_documents(documents=documents, embedding=embeddings)
+    
+    return vectorstore
+
+def answer_query(query, vectorstore):
+    retriever = vectorstore.as_retriever()
+    search_results = retriever.get_relevant_documents(query)
+    
+    context = " ".join([doc.page_content for doc in search_results])
+    
+    generator = pipeline("text2text-generation", model="google/flan-t5-base")
+
+    prompt = f"Based on this text: {context}\nAnswer this question: {query}"
+    generated_text = generator(prompt, max_length=100)
+
+    return generated_text[0]['generated_text']

requirements.txt

@@ -0,0 +1,9 @@
+gradio==3.50.2
+langchain==0.1.14
+langchain_community==0.0.30
+langchain-core==0.1.37
+transformers==4.36.2
+faiss-cpu==1.7.4
+sentence-transformers==2.5.1
+huggingface_hub>=0.20.0
+pydantic<2.0