Upload 3 files

app.py

@@ -1,49 +1,70 @@
+#Allows you to use Streamlit, a framework for building interactive web applications.
+#It provides functions for creating UIs, displaying data, and handling user inputs.
 import streamlit as st
-from langchain.chat_models import ChatOpenAI
-from langchain.schema import AIMessage, HumanMessage, SystemMessage
 
-# From here down is all the StreamLit UI.
-st.set_page_config(page_title="LangChain Demo", page_icon=":robot:")
-st.header("Hey, I'm your Chat GPT")
 
+#This module provides a way to interact with the operating system, such as accessing environment variables, working with files
+#and directories, executing shell commands, etc
+import os
 
-if "sessionMessages" not in st.session_state:
-    st.session_state.sessionMessages = [
-        SystemMessage(content="You are a helpful assistant.")
-    ]
+#Helps us generate embeddings
+#An embedding is a vector (list) of floating point numbers. The distance between two vectors measures their relatedness. 
+#Small distances suggest high relatedness and large distances suggest low relatedness.
+from langchain.embeddings import OpenAIEmbeddings
 
 
-def load_answer(question):
+#FAISS is an open-source library developed by Facebook AI Research for efficient similarity search and clustering of large-scale datasets, particularly with high-dimensional vectors. 
+#It provides optimized indexing structures and algorithms for tasks like nearest neighbor search and recommendation systems.
+from langchain.vectorstores import FAISS
 
-    st.session_state.sessionMessages.append(HumanMessage(content=question))
 
-    assistant_answer = chat(st.session_state.sessionMessages)
+#load_dotenv() is a function that loads variables from a .env file into environment variables in a Python script. 
+#It allows you to store sensitive information or configuration settings separate from your code
+#and access them within your application.
+from dotenv import load_dotenv
 
-    # st.session_state.sessionMessages.append(
-    #     AIMessage(content=assistant_answer.content)
-    # )
 
-    st.session_state.sessionMessages.append(assistant_answer)
-    assistant_answer
+load_dotenv()
 
-    return assistant_answer
 
+#By using st.set_page_config(), you can customize the appearance of your Streamlit application's web page
+st.set_page_config(page_title="Educate Kids", page_icon=":robot:")
+st.header("Hey, Ask me something & I will give out similar things")
 
-def get_text():
-    input_text = st.text_input("You: ", key=input)
-    return input_text
+#Initialize the OpenAIEmbeddings object
+embeddings = OpenAIEmbeddings()
 
+#The below snippet helps us to import CSV file data for our tasks
+from langchain.document_loaders.csv_loader import CSVLoader
+loader = CSVLoader(file_path='myData.csv', csv_args={
+    'delimiter': ',',
+    'quotechar': '"',
+    'fieldnames': ['Words']
+})
 
-chat = ChatOpenAI(temperature=0)
+#Assigning the data inside the csv to our variable here
+data = loader.load()
 
+#Display the data
+print(data)
 
-user_input = get_text()
-submit = st.button("Generate")
+db = FAISS.from_documents(data, embeddings)
+
+#Function to receive input from user and store it in a variable
+def get_text():
+    input_text = st.text_input("You: ", key= input)
+    return input_text
 
-if submit:
 
-    answer = load_answer(user_input)
-    response = answer.content
-    st.subheader("Answer:")
+user_input=get_text()
+submit = st.button('Find similar Things')  
+
+if submit:
+    
+    #If the button is clicked, the below snippet will fetch us the similar text
+    docs = db.similarity_search(user_input)
+    print(docs)
+    st.subheader("Top Matches:")
+    st.text(docs[0])
+    st.text(docs[1].page_content)
 
-    st.write(response, key=1)

myData.csv

@@ -0,0 +1,12 @@
+Words
+Elephant
+Lion
+Tiger
+Dog
+Cricket
+Footbal
+Tennis
+Basketball
+Apple
+Orange
+Banana

requirements.txt

@@ -1,5 +1,6 @@
-streamlit
 langchain
+streamlit
 openai
-streamlit-chat
-python-dotenv
+tiktoken
+python-dotenv
+faiss-cpu