File size: 5,140 Bytes
c884422
 
dfe7961
9c97c75
5839dd5
9c97c75
5839dd5
9c97c75
 
 
 
 
5839dd5
c884422
dfe7961
9c97c75
5839dd5
4ddb082
89cfcf5
5839dd5
dfe7961
9c97c75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7339830
9c97c75
7339830
9c97c75
 
 
 
 
 
97131a6
 
9c97c75
97131a6
9c97c75
 
97131a6
9c97c75
 
 
97131a6
 
 
9c97c75
 
 
 
 
 
 
 
7339830
 
 
 
97131a6
 
9c97c75
c884422
 
163c830
c884422
 
 
9c97c75
 
c884422
9c97c75
5839dd5
9c97c75
 
c884422
 
5839dd5
c884422
5839dd5
 
 
9c97c75
 
 
 
 
 
c884422
9c97c75
 
 
 
c884422
9c97c75
 
c884422
9c97c75
e3e6646
9c97c75
e3e6646
 
 
9c97c75
 
 
 
 
 
 
 
 
 
 
e3e6646
 
 
 
 
 
 
 
 
 
 
9c97c75
 
 
 
5839dd5
 
 
7339830
9c97c75
7339830
9c97c75
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154


import streamlit as st
from langchain.prompts import PromptTemplate
from langchain_community.llms import CTransformers
from src.helper import download_hf_embeddings, text_split, download_hf_model
from langchain_community.vectorstores import Pinecone as LangchainPinecone
import os
from dotenv import load_dotenv
from src.prompt import prompt_template
from langchain.chains import RetrievalQA
import time
from pinecone import Pinecone
from tqdm.auto import tqdm

# Load environment variables
load_dotenv()

PINECONE_API_KEY = os.getenv('PINECONE_API_KEY')
index_name = "medicure-chatbot"

# Set page configuration
st.set_page_config(page_title="Medical Chatbot", page_icon="πŸ₯", layout="wide")

# Custom CSS for styling
st.markdown("""
<style>
    .stApp {
        background-color: #f0f8ff;
    }
    .stButton>button {
        background-color: #4CAF50;
        color: white;
        border-radius: 20px;
        border: none;
        padding: 10px 20px;
        transition: all 0.3s ease;
    }
    .stButton>button:hover {
        background-color: #333;
        transform: scale(1.05);
        color:#fff;
    }
    .footer {
        position: fixed;
        left: 0;
        bottom: 0;
        width: 100%;
        background-color: #f0f8ff ;
        color: #333;
        text-align: center;
        
    }
    .social-icons a {
        color: #333;
        margin: 0 10px;
        font-size: 24px;
    }
    .social-icons a>social-icons a:hover {
        color: #4CAF50;
    }
</style>
""", unsafe_allow_html=True)

# Initialize session state for chat history
if 'chat_history' not in st.session_state:
    st.session_state.chat_history = []

# Header
st.title("πŸ₯ Medicure RAG Chatbot")

# Display welcome message
st.write("Welcome to Medicure Chatbot! Ask any medical question and I'll do my best to help you.")
st.write("#### Built with πŸ€— Ctransformers, Langchain, and Pinecone VectorDB. Powered by Metal-llama2-7b-chat quantized LLM")
st.write("##### Resource Used πŸ“– : The Gale Encyclopedia of Medicine ")

# Parameters section
st.sidebar.header("Parameters")
k_value = st.sidebar.slider("Number of relevant documents (k)", min_value=1, max_value=3, value=2)
max_new_tokens = st.sidebar.slider("Max new tokens", min_value=64, max_value=1024, value=512)
temperature = st.sidebar.slider("Temperature", min_value=0.1, max_value=1.0, value=0.8, step=0.1)

# Initialize the chatbot components
@st.cache_resource
def initialize_chatbot(k, max_tokens, temp):
    embeddings = download_hf_embeddings()
    model_path = "TheBloke/Llama-2-7B-Chat-GGML"
    llm = CTransformers(model=model_path,
                        model_type="llama",
                        config={'max_new_tokens': max_tokens,
                                'temperature': temp})
    
    # initialize pinecone
    pc = Pinecone(api_key=PINECONE_API_KEY)
    index = pc.Index(index_name)

    PROMPT = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
    chain_type_kwargs = {"prompt": PROMPT}
    docsearch = LangchainPinecone(index, embeddings.embed_query, "text")
    qa = RetrievalQA.from_chain_type(
        llm=llm,
        chain_type="stuff",
        retriever=docsearch.as_retriever(search_kwargs={'k': k}),
        return_source_documents=True,
        chain_type_kwargs=chain_type_kwargs)
    return qa

qa = initialize_chatbot(k_value, max_new_tokens, temperature)

# Chat interface
user_input = st.text_input("Ask your question:")
if st.button("Send", key="send"):

    if user_input:
        with st.spinner("Thinking..."):
            result = qa({"query": user_input})
            response = result["result"]
        st.session_state.chat_history.append(("You", user_input))
        st.session_state.chat_history.append(("Bot", response))

# Display chat history
st.subheader("Chat History")
for role, message in st.session_state.chat_history:
    if role == "You":
        st.markdown(f"**You:** {message}")
    else:
        st.markdown(f"**Bot:** {message}")


# Animated loading for visual appeal
def load_animation():
    with st.empty():
        for i in range(3):
            for j in ["β‹…", "β‹…β‹…", "β‹…β‹…β‹…", "β‹…β‹…β‹…β‹…"]:
                st.write(f"Loading{j}")
                time.sleep(0.2)
            st.write("")


# Footer with social links
st.markdown("""
<div class="footer">
    <div class="social-icons">
        <a href="https://github.com/4darsh-Dev" target="_blank"><i class="fab fa-github"></i></a>
        <a href="https://linkedin.com/in/adarsh-maurya-dev" target="_blank"><i class="fab fa-linkedin"></i></a>
        <a href="https://adarshmaurya.onionreads.com" target="_blank"><i class="fas fa-globe"></i></a>
        <a href="https://www.kaggle.com/adarshm09" target="_blank"><i class="fab fa-kaggle"></i></a>
    </div>
    <p> <p style="text-align:center;">Made with ❀️ by <a href="https://www.adarshmaurya.onionreads.com">Adarsh Maurya</a></p> </p>
</div>
""", unsafe_allow_html=True)

# Load Font Awesome for icons
st.markdown('<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.15.1/css/all.min.css">', unsafe_allow_html=True)