Create app.py

app.py

@@ -0,0 +1,89 @@
+pip install -qU langchain-community faiss-cpu faiss-gpu langchain-openai sentence_transformers gradio
+
+import faiss
+from langchain_community.docstore.in_memory import InMemoryDocstore
+from langchain_community.vectorstores import FAISS
+from langchain_openai import OpenAIEmbeddings
+import os
+import pandas as pd
+from uuid import uuid4
+from langchain_core.documents import Document
+import numpy as np
+from sentence_transformers import SentenceTransformer
+from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter
+from langchain.chains import RetrievalQA
+from langchain.llms import OpenAI
+from langchain_core.prompts import ChatPromptTemplate
+from langchain import PromptTemplate
+import gradio as gr
+
+df = pd.read_csv('news_paper-Cleaned.csv', encoding='utf-8', on_bad_lines='skip')
+
+os.environ["OPENAI_API_KEY"] = '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'
+embeddings = OpenAIEmbeddings(model="text-embedding-3-large")
+
+index = faiss.IndexFlatL2(len(embeddings.embed_query("hello world")))
+
+vector_store = FAISS(
+    embedding_function=embeddings,
+    index=index,
+    docstore=InMemoryDocstore(),
+    index_to_docstore_id={},
+)
+
+documents = [{
+    'title': row['title'],
+    'author': row['author'],
+    'description': row['description'],
+    'full_text' : row['full_text']
+
+}
+    for _, row in df.iterrows()]
+
+full_text = [Document(
+    page_content=str(doc),
+    metadata={"source": "news"},
+) for doc in documents]
+
+text_splitter = RecursiveCharacterTextSplitter(
+    # Set a really small chunk size, just to show.
+    chunk_size=1000,
+    chunk_overlap=100,
+    length_function=len,
+    is_separator_regex=False,
+)
+
+text_split = text_splitter.split_documents(full_text)
+
+uuids = [str(uuid4()) for _ in range(len(text_split))]
+
+vector_store.add_documents(documents=text_split, ids=uuids)
+
+retriever = vector_store.as_retriever(search_type="mmr", search_kwargs={"k": 10})
+
+def questions(query):
+
+    template = """
+    You are a helpful assistant that that can answer questions about specific data.
+    You have answer only from this Context.
+    You will receive 10 Answer return all and spilt between them by new line.
+
+    Question: {question}
+    Context: {context}
+    Answer: 
+    """
+
+
+    PROMPT = PromptTemplate(template=template, input_variables=['question', 'context'])
+
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=OpenAI(), 
+        retriever=retriever,
+        chain_type_kwargs={"prompt": PROMPT},
+    )
+
+    return qa_chain({"query": query})['result']
+
+
+demo = gr.Interface(fn=questions, inputs="text", outputs="text")
+demo.launch()