app.py

from google.colab import drive
drive.mount("/content/drive")
!pip install langchain sentence-transformers chromadb llama-cpp-python langchain_community pypdf
from langchain_community.document_loaders import PyPDFDirectoryLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_community.embeddings import SentenceTransformerEmbeddings
from langchain.vectorstores import Chroma
from langchain_community.llms import LlamaCpp
from langchain.chains import RetrievalQA, LLMChain
loader=PyPDFDirectoryLoader("/content/drive/MyDrive/BioMistral/Data")
docs=loader.load()
text_splitter=RecursiveCharacterTextSplitter(chunk_size=300, chunk_overlap=50)
chunks=text_splitter.split_documents(docs)
import os
import os
os.environ["HUGGINGFACEHUB_API_TOKEN"] = os.getenv("HUGGINGFACEHUB_API_TOKEN")

embeddings = SentenceTransformerEmbeddings(model_name="NeuML/pubmedbert-base-embeddings")
vectorstore = Chroma.from_documents(chunks, embeddings)
query="What are the major risk factors of heart disease?"
search_results=vectorstore.similarity_search(query)
search_results
retriever=vectorstore.as_retriever(search_kwargs={"k":5})
retriever.get_relevant_documents(query)
llm=LlamaCpp(
    model_path="/content/drive/MyDrive/BioMistral/BioMistral-7B.Q4_K_M.gguf",
    temperature=0.2,
    max_tokens=2048,
    top_p=1
)
template="""
<|context|>
You are an Medical Assistant that follows the instruction and generate the accurate response based on the query and the context provided.
Please be truthful and give direct answers.
</s>
<|user|>
{query}
</s>
<|assistant|>
"""
from langchain.schema.runnable import RunnablePassthrough
from langchain.schema.output_parser import StrOutputParser
from langchain.prompts import ChatPromptTemplate
prompt=ChatPromptTemplate.from_template(template)
rag_chain=(
    {"context":retriever,"query":RunnablePassthrough()}
    | prompt
    | llm
    | StrOutputParser()
  )
response=rag_chain.invoke("query")
response
import sys
while True:
  user_input=input(f"Input query: ")
  if user_input=='exit':
    print("Exiting...")
    sys.exit()
  if user_input=="":
    continue
  result=rag_chain.invoke(user_input)
  print("Answer: ",result)
!pip install gradio
import gradio as gr

# Define a function to handle queries
def chatbot_ui(user_query):
    if not user_query.strip():
        return "Please enter a valid query."
    try:
        result = rag_chain.invoke(user_query)
        return result
    except Exception as e:
        return f"Error: {str(e)}"

# Create the Gradio interface
interface = gr.Interface(
    fn=chatbot_ui,  # Function to process the query
    inputs=gr.Textbox(label="Enter your medical query:", placeholder="Ask a medical question here..."),
    outputs=gr.Textbox(label="Chatbot Response"),
    title="Medical Assistant Chatbot",
    description="A chatbot made for heart patients.",
    examples=[
        ["What are the symptoms of diabetes?"],
        ["Explain the risk factors of heart disease."],
        ["How can I reduce cholesterol levels naturally?"],
    ]
)

# Launch the Gradio interface
interface.launch(share=True)