Upload 5 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+unsloth.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text

a.py

@@ -0,0 +1,166 @@
+import streamlit as st
+import pandas as pd
+from fuzzywuzzy import process
+from langchain_community.llms import LlamaCpp
+from langchain_core.callbacks import StreamingStdOutCallbackHandler
+from langchain_core.prompts import PromptTemplate
+
+# Load the CSV files into DataFrames with Windows-1252 encoding
+df = pd.read_csv('location.csv', encoding='Windows-1252')
+df2 = pd.read_csv('train.csv')
+
+# Initialize the LlamaCpp model
+llm = LlamaCpp(
+    model_path="unsloth.Q5_K_M.gguf",
+    temperature=0.01,
+    max_tokens=500,
+    top_p=3,
+    callbacks=[StreamingStdOutCallbackHandler()],
+    verbose=False,
+    stop=["###"]
+)
+
+# Define the prompt template
+template = """Below is an instruction that describes a task, paired with an input that provides further context. Write a lengthy detailed response that appropriately completes the request.
+
+### Instruction:
+{instruction}
+
+### Input:
+{input}
+
+### Response:
+{response}"""
+
+prompt = PromptTemplate.from_template(template)
+
+
+# Function to find the best matching context based on user input
+def find_best_match(query):
+    questions = df2['Question'].tolist()
+    contexts = df2['Context'].tolist()
+
+    # Find the best match
+    best_match = process.extractOne(query, questions)
+    if best_match:
+        index = questions.index(best_match[0])
+        return contexts[index]
+    return "No relevant information found."
+
+
+# Function to truncate response at the nearest full stop
+def truncate_at_full_stop(text, max_length=500):
+    if len(text) <= max_length:
+        return text
+
+    truncated = text[:max_length]
+    print(f"Truncated text: {truncated}")
+
+    last_period = truncated.rfind('.')
+    print(f"Last period index: {last_period}")
+
+    if last_period != -1:
+        return truncated[:last_period + 1]
+
+    return truncated
+
+
+# Initialize session state for selected service, chat history, and AI history
+if 'selected_service' not in st.session_state:
+    st.session_state.selected_service = "Home"
+if 'chat_history' not in st.session_state:
+    st.session_state.chat_history = []
+if 'history' not in st.session_state:
+    st.session_state.history = []
+if 'input' not in st.session_state:
+    st.session_state['input'] = ''
+
+# Sidebar for selecting services
+with st.sidebar:
+    st.title("Select the Service")
+
+    # Create buttons for each service
+    if st.button('Medicine Services'):
+        st.session_state.selected_service = "Medicine Services"
+
+    if st.button('Kendra Locator'):
+        st.session_state.selected_service = "Kendra Locator"
+
+    if st.button('Assistant'):
+        st.session_state.selected_service = "Assistant"
+
+# Main content area based on selected service
+if st.session_state.selected_service == "Home":
+    st.title("Welcome to Medical Service Center")
+    st.write("Explore the options in the sidebar to get started.")
+
+elif st.session_state.selected_service == "Medicine Services":
+    st.title("Medicine Services")
+
+    # Display chat history
+    for chat in st.session_state.chat_history:
+        st.write(f"**User:** {chat['user']}")
+        st.write(f"**Bot:** {chat['bot']}")
+
+    # User input section
+    user_input = st.text_input("Enter medicine:")
+
+    # Handle input when the "Send" button is clicked
+    if st.button('Send'):
+        if user_input:
+            response = find_best_match(user_input)
+            st.session_state.chat_history.append({"user": user_input, "bot": response})
+
+elif st.session_state.selected_service == "Kendra Locator":
+    st.title("Kendra Locator")
+    display_option = st.selectbox("Select:", ["Address", "Email"])
+    pin_code_input = st.text_input("Enter Pin Code:")
+
+    if st.button("Locate"):
+        if pin_code_input:
+            result = df[df['Pin'].astype(str) == pin_code_input]
+            if not result.empty:
+                if display_option == "Address":
+                    st.write(f"Address: {result['Address'].values[0]}")
+                elif display_option == "Email":
+                    st.write(f"Email: {result['Email'].values[0]}")
+            else:
+                st.write("No results found.")
+        else:
+            st.write("Please enter a pin code.")
+
+elif st.session_state.selected_service == "Assistant":
+    st.title("Query Assistance")
+
+    # Display AI chat history
+    for chat in st.session_state.history:
+        st.write(f"**Medicine Query:** {chat['user']}")
+        st.write(f"**Chatbot:** {chat['bot']}")
+
+
+    # Function to handle user input
+    def handle_input():
+        user_input = st.session_state['input']
+        if user_input:
+            # Format the prompt
+            formatted_prompt = prompt.format(
+                instruction="You are an all-knowing Medical AI. Provide detailed responses to only medicine-related queries.",
+                input=user_input,
+                response=""  # Leave this blank for generation!
+            )
+
+            # Generate response
+            response = llm.invoke(formatted_prompt)
+
+            # Truncate response if necessary
+            truncated_response = truncate_at_full_stop(response)
+
+            # Update the chat history
+            st.session_state.history.append({"user": user_input, "bot": truncated_response})
+
+            # Clear the input box
+            st.session_state['input'] = ''
+
+
+    # Persistent text input at the top
+    st.text_input("Enter Medicine Name:", key="input", on_change=handle_input)

unsloth.Q5_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56ec872ac6e16e492c06073affb524dbc121d4b0d3d906edbbfe219231b1bfc9
+size 5732987264