diff --git a/README.md b/README.md
index f58a8d6..1d75346 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,9 @@
# Personal AI Assistant: Your Web and PDF Guide
+## How to run on local desktop
+* Install correct packages
+* Run in CMD
+* Navigate to location of the notebook: ex. cd C:\Users\felixstuyck\Documents
+* streamlit run "Ai assistent script.py"
diff --git a/Test FAISS package.py b/Test FAISS package.py
new file mode 100644
index 0000000..5c22075
--- /dev/null
+++ b/Test FAISS package.py
@@ -0,0 +1,46 @@
+import getpass
+import os
+import configparser
+from langchain_community.document_loaders import TextLoader
+from langchain_community.vectorstores import FAISS
+from langchain_openai import OpenAIEmbeddings
+from langchain_text_splitters import CharacterTextSplitter
+
+# Get open AI key
+config_path = r"C:\Users\felixstuyck\OneDrive - Finvision\Documenten\Python\AI assistant config.ini"
+config = configparser.ConfigParser()
+config.read(config_path)
+OPEN_AI_KEY = config['API_KEY_Assistent']['API_KEY']
+os.environ["OPENAI_API_KEY"] = getpass.getpass(OPEN_AI_KEY)
+
+# Load and process the document
+loader = TextLoader(R"C:\Users\felixstuyck\OneDrive - Finvision\Documenten\Stow stored procedure.txt")
+documents = loader.load()
+text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
+docs = text_splitter.split_documents(documents)
+
+# Generate embeddings and create FAISS index
+embeddings = OpenAIEmbeddings()
+db = FAISS.from_documents(docs, embeddings)
+print(f"Number of documents in the index: {db.index.ntotal}")
+
+# Query the vectorstore
+query = "What did the president say about Ketanji Brown Jackson"
+docs = db.similarity_search(query)
+print(f"Top result content: {docs[0].page_content}")
+
+# Use as retriever
+retriever = db.as_retriever()
+docs = retriever.invoke(query)
+print(f"Top result content (retriever): {docs[0].page_content}")
+
+# Similarity search with score
+docs_and_scores = db.similarity_search_with_score(query)
+for doc, score in docs_and_scores:
+ print(f"Content: {doc.page_content}, Score: {score}")
+
+# Save and load FAISS index
+db.save_local("faiss_index")
+new_db = FAISS.load_local("faiss_index", embeddings)
+docs = new_db.similarity_search(query)
+print(f"Top result content (loaded index): {docs[0].page_content}")
diff --git a/main.py b/main.py
index 97b87d4..7b4095a 100644
--- a/main.py
+++ b/main.py
@@ -1,107 +1,47 @@
+import getpass
import os
-import streamlit as st
-import pickle
-import PyPDF2
-from PyPDF2 import PdfReader
-from langchain.document_loaders import PyPDFLoader
-from langchain.chains.summarize import load_summarize_chain
-import tempfile
-import time
-import langchain
-from langchain import OpenAI
-from langchain.chains import RetrievalQAWithSourcesChain
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from langchain.document_loaders import UnstructuredURLLoader
-from langchain.embeddings import OpenAIEmbeddings
-from langchain.vectorstores import FAISS
-from langchain.chains.question_answering import load_qa_chain
-
-
-OPENAI_API_KEY ='User Your Open AI API KEY'
-url_file_path = "url_faiss_store_openai.pkl"
-
-# Streamlit setup
-st.set_page_config(
- page_title="Personal AI Assistant",
- layout="wide",
- initial_sidebar_state="expanded"
-)
-st.markdown("🤖 Personal AI Assistant
", unsafe_allow_html=True)
-st.sidebar.markdown("Assistant Console
", unsafe_allow_html=True)
-
-# ---- URL Loading & Embedding ----
-num_links = st.sidebar.slider("How many links do you want to input?", min_value=1, max_value=5, value=1)
-urls = [st.sidebar.text_input(f"URL {i+1}", key=f"url{i}") for i in range(num_links)]
-if urls:
- loader = UnstructuredURLLoader(urls=urls)
- data = loader.load()
- text_splitter = RecursiveCharacterTextSplitter(separators=["\n\n", "\n", "."], chunk_size=1000)
- url_docs = text_splitter.split_documents(data)
- if url_docs:
- embeddings = OpenAIEmbeddings(openai_api_key=OPENAI_API_KEY)
- url_vectorindex_openai = FAISS.from_documents(url_docs, embeddings)
- with open(url_file_path, "wb") as f:
- pickle.dump(url_vectorindex_openai, f)
-
-# ---- PDF Loading & Embedding ----
-uploaded_file = st.sidebar.file_uploader("Upload a PDF file", type=['pdf'])
-if uploaded_file:
- pdf_reader = PdfReader(uploaded_file)
- pdf_text = ""
- for page in pdf_reader.pages:
- pdf_text += page.extract_text()
- text_splitter = RecursiveCharacterTextSplitter(separators=["\n\n", "\n", "."], chunk_size= 500)
- pdf_docs = text_splitter.split_text(pdf_text)
- if pdf_docs:
- embeddings = OpenAIEmbeddings(openai_api_key=OPENAI_API_KEY)
- pdf_vectors = FAISS.from_texts(pdf_docs, embeddings)
-
-
-# ---- Query Interface ----
-llm = OpenAI(temperature=0.9, max_tokens=500, openai_api_key=OPENAI_API_KEY)
-data_source = st.selectbox("What do you want to inquire about?", ["URL", "PDF"])
-
-if data_source == "URL":
- query_url = st.text_input('Ask your question about URLs:')
- if query_url:
- if os.path.exists(url_file_path): # Ensure URL database exists
- with open(url_file_path, "rb") as f:
- vectorstore = pickle.load(f)
- chain = RetrievalQAWithSourcesChain.from_llm(llm=llm, retriever=vectorstore.as_retriever())
- result = chain({"question": query_url}, return_only_outputs=True)
- st.header("Answer based on URLs:")
- st.subheader(result['answer'])
-
-elif data_source == "PDF":
- query_pdf = st.text_input('Ask your question about PDFs:')
- if query_pdf:
- docs = pdf_vectors.similarity_search(query_pdf)
-
- chain = load_qa_chain(llm, chain_type="stuff")
- response = chain.run(input_documents=docs, question=query_pdf)
-
- st.write(response)
-
- if st.button("Summarize PDF"):
- def summarize_pdfs_from_folder(pdfs_folder):
- summaries = []
- for pdf_file in pdfs_folder:
- with tempfile.NamedTemporaryFile(delete=False) as temp_file:
- temp_path = temp_file.name
- temp_file.write(pdf_file.getvalue())
- loader = PyPDFLoader(temp_path)
- docs = loader.load_and_split()
- chain = load_summarize_chain(llm, chain_type="map_reduce")
- summary = chain.run(docs)
- summaries.append(summary)
- os.remove(temp_path)
- return summaries
-
- summaries = summarize_pdfs_from_folder([uploaded_file])
- for summary in summaries:
- st.write(summary)
-
-
-
-
-
+import configparser
+
+# Get open AI key
+config_path = r"C:\Users\felixstuyck\OneDrive - Finvision\Documenten\Python\AI assistant config.ini"
+config = configparser.ConfigParser()
+config.read(config_path)
+OPEN_AI_KEY = config['API_KEY_Assistent']['API_KEY']
+os.environ["OPENAI_API_KEY"] = getpass.getpass(OPEN_AI_KEY)
+
+from langchain_community.document_loaders import TextLoader
+from langchain_community.vectorstores import FAISS
+from langchain_openai import OpenAIEmbeddings
+from langchain_text_splitters import CharacterTextSplitter
+
+# Load and process the document
+loader = TextLoader(R"C:\Users\felixstuyck\OneDrive - Finvision\Documenten\Stow stored procedure.txt")
+documents = loader.load()
+text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
+docs = text_splitter.split_documents(documents)
+
+# Generate embeddings and create FAISS index
+embeddings = OpenAIEmbeddings()
+db = FAISS.from_documents(docs, embeddings)
+print(f"Number of documents in the index: {db.index.ntotal}")
+
+# Query the vectorstore
+query = "What did the president say about Ketanji Brown Jackson"
+docs = db.similarity_search(query)
+print(f"Top result content: {docs[0].page_content}")
+
+# Use as retriever
+retriever = db.as_retriever()
+docs = retriever.invoke(query)
+print(f"Top result content (retriever): {docs[0].page_content}")
+
+# Similarity search with score
+docs_and_scores = db.similarity_search_with_score(query)
+for doc, score in docs_and_scores:
+ print(f"Content: {doc.page_content}, Score: {score}")
+
+# Save and load FAISS index
+db.save_local("faiss_index")
+new_db = FAISS.load_local("faiss_index", embeddings)
+docs = new_db.similarity_search(query)
+print(f"Top result content (loaded index): {docs[0].page_content}")
\ No newline at end of file