main functionality

app.py

@@ -1,63 +1,68 @@
 import streamlit as st
-from langchain.memory import ConversationBufferMemory
-from langchain.chains import LLMChain
-from langchain_community.llms import HuggingFaceHub
-from langchain.prompts import PromptTemplate
-from dotenv import load_dotenv
-import time
-
-def get_response(model, query):
-    prompt_template = PromptTemplate(
-            template="I have a question about my health. {user_question}",
-            input_variables=["user_question"]
-        )
-    # get the response
-    memory = ConversationBufferMemory(memory_key="messages", return_messages=True)
-    print(memory)
-    conversation_chain = LLMChain(
-        llm=model,
-        prompt=prompt_template,
-        # retriever=vectorstore.as_retriever(),
-        memory=memory)
-    response = conversation_chain.invoke(query)
-    answer = response["text"]
-    if "\n\n" in answer:
-        answer = answer.split("\n\n", 1)[1]
-    return answer
+import openai
 
+# Function to interact with GPT-3 model
+def ask_gpt3_personalized(prompt, height_cm, weight_kg, age):
+    response = openai.ChatCompletion.create(
+        model="gpt-3.5-turbo",
+        messages=[
+            {"role": "system", "content": "You are a chatbot that will answer queries related to fitness and nutrition. The chatbot should understand questions about workout routines, dietary advice, and general fitness tips. Chatbot will offer personalized workout and diet plans based on user inputs such as body type, fitness goals, and dietary restrictions. this is the chat history use this as a reference to answer the queries"+st.session_state.messages},
+            {"role": "user", "content": "my height is " + str(height_cm) + " cm, my weight is " + str(weight_kg) + " kg, and I am " + str(age) + " years old."+prompt},
+        ]
+    )
+    return response['choices'][0]['message']['content']
+
+def ask_gpt3(prompt):
+    response = openai.ChatCompletion.create(
+        model="gpt-3.5-turbo",
+        messages=[
+            {"role": "system", "content": "You are a chatbot that will answer queries related to fitness and nutrition. The chatbot should understand questions about workout routines, dietary advice, and general fitness tips. Chatbot will offer personalized workout and diet plans based on user inputs such as body type, fitness goals, and dietary restrictions. this is the chat history use this as a reference to answer the queries"+st.session_state.messages},
+            {"role": "user", "content": prompt},
+        ]
+    )
+    return response['choices'][0]['message']['content']
+
+# Main function to run the Streamlit app
 def main():
-    st.title("Health Chatbot")
-    # load the environment variables    
-    load_dotenv()
-    print("Loading LLM from HuggingFace")
-    with st.spinner('Loading LLM from HuggingFace...'):
-        # llm = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.7, "max_new_tokens":1028, "top_p":0.95})
-
-        llm = HuggingFaceHub(repo_id="epfl-llm/meditron-70b", model_kwargs={"temperature":0.7, "max_new_tokens":1028, "top_p":0.95})
+    st.title("Fitness Chatbot")
+    
+    # add a sibebar that can take in user input for the chatbot prompt and the API key
+    st.sidebar.title("Personal Information")
+    openai.api_key = st.sidebar.text_input("Your OpenAI API Key here", "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")
+    height_cm = st.sidebar.number_input("Height (cm)", 0, 300)
+    weight_kg = st.sidebar.number_input("Weight (kg)", 0, 300)
+    age = st.sidebar.number_input("Age", 0, 100)
+    
     
+    # Initialize conversation history
     if "messages" not in st.session_state:
         st.session_state.messages = []
-        
-        
-    if st.button("Clear Chat"):
-        st.session_state.messages = []
-    
-    for message in st.session_state.messages:
-        if message["role"] == "user":
-            st.chat_message("user").markdown(message["content"])
+
+    # User input
+    user_input = st.chat_input("Ask something")
+
+    # If user input is not empty
+    if user_input:
+        # Add user input to conversation history
+        st.session_state.messages.append({"role": "user", "content": user_input})
+
+        # Get chatbot response
+        if height_cm and weight_kg and age:
+            chatbot_response = ask_gpt3_personalized(user_input, height_cm, weight_kg, age)
         else:
-            st.chat_message("bot").markdown(message["content"])
-    
-    user_prompt = st.chat_input("ask a question", key="user")
-    if user_prompt:
-        st.chat_message("user").markdown(user_prompt)
-        st.session_state.messages.append({"role": "user", "content": user_prompt})
-        with st.spinner('Thinking...'):
-            start_time = time.time()
-            response = get_response(llm, user_prompt)
-            st.write("Response Time: ", time.time() - start_time)
-        st.chat_message("bot").markdown(response)
-        st.session_state.messages.append({"role": "bot", "content": response})
-        
+            chatbot_response = ask_gpt3(user_input)
+
+        # Add chatbot response to conversation history
+        st.session_state.messages.append({"role": "assistant", "content": chatbot_response})
+
+        # Display conversation history
+        for message in st.session_state.messages:
+            with st.chat_message(message["role"]):
+                st.markdown(message["content"])
+
+        # Clear the chat input
+        st.session_state["chat_input"] = ""
+
+# Run the main function
 if __name__ == "__main__":
     main()

app2.py

@@ -0,0 +1,63 @@
+import streamlit as st
+from langchain.memory import ConversationBufferMemory
+from langchain.chains import LLMChain
+from langchain_community.llms import HuggingFaceHub
+from langchain.prompts import PromptTemplate
+from dotenv import load_dotenv
+import time
+
+def get_response(model, query):
+    prompt_template = PromptTemplate(
+            template="I have a question about my health. {user_question}",
+            input_variables=["user_question"]
+        )
+    # get the response
+    memory = ConversationBufferMemory(memory_key="messages", return_messages=True)
+    print(memory)
+    conversation_chain = LLMChain(
+        llm=model,
+        prompt=prompt_template,
+        # retriever=vectorstore.as_retriever(),
+        memory=memory)
+    response = conversation_chain.invoke(query)
+    answer = response["text"]
+    if "\n\n" in answer:
+        answer = answer.split("\n\n", 1)[1]
+    return answer
+
+def main():
+    st.title("Health Chatbot")
+    # load the environment variables    
+    load_dotenv()
+    print("Loading LLM from HuggingFace")
+    with st.spinner('Loading LLM from HuggingFace...'):
+        llm = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.7, "max_new_tokens":1028, "top_p":0.95})
+
+        # llm = HuggingFaceHub(repo_id="ajdev/falcon_medical", model_kwargs={"temperature":0.7, "max_new_tokens":250, "top_p":0.95})
+    
+    if "messages" not in st.session_state:
+        st.session_state.messages = []
+        
+        
+    if st.button("Clear Chat"):
+        st.session_state.messages = []
+    
+    for message in st.session_state.messages:
+        if message["role"] == "user":
+            st.chat_message("user").markdown(message["content"])
+        else:
+            st.chat_message("bot").markdown(message["content"])
+    
+    user_prompt = st.chat_input("ask a question", key="user")
+    if user_prompt:
+        st.chat_message("user").markdown(user_prompt)
+        st.session_state.messages.append({"role": "user", "content": user_prompt})
+        with st.spinner('Thinking...'):
+            start_time = time.time()
+            response = get_response(llm, user_prompt)
+            st.write("Response Time: ", time.time() - start_time)
+        st.chat_message("bot").markdown(response)
+        st.session_state.messages.append({"role": "bot", "content": response})
+        
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -1,4 +1,5 @@
 streamlit
 langchain
 langchain-community
-InstructorEmbedding
+InstructorEmbedding
+openai==0.28