Update app.py

app.py

@@ -148,213 +148,43 @@ placeholders = [
     'Example: Does this resource provide guidance on implementation fidelity'
 ]
 
-# Select a random placeholder from the list
-if 'placeholder' not in st.session_state:
-    st.session_state.placeholder = random.choice(placeholders)
-    
-q = st.text_input(label='Ask a question or make a request ', value='', placeholder=st.session_state.placeholder)
-# q = st.text_input(label='Ask a question or make a request ', value='')
-    
-if q:
-    with st.spinner('Thinking...'):
-        answer = ask_with_memory(vector_store, q, st.session_state.history)
-    
-    # Display the response in a text area
-    # st.text_area('Response: ', value=answer, height=400, key="response_text_area")
-    import time
-    import random
-
-    def stream_data():
-        for word in answer.split(" "):
-            yield word + " "
-            # time.sleep(0.02)
-            time.sleep(random.uniform(0.03, 0.08))
-
-    st.write(stream_data)
-    
-    st.success('Powered by MTSS GPT. AI can make mistakes. Consider checking important information.')
-
-    # Prepare chat history text for display
-    # history_text = "\n\n".join(f"Q: {entry[0]}\nA: {entry[1]}" for entry in st.session_state.history)
-    # Prepare chat history text for display in reverse order
-    history_text = "\n\n".join(f"Q: {entry[0]}\nA: {entry[1]}" for entry in reversed(st.session_state.history))
-
-    # Display chat history
-    st.text_area('Chat History', value=history_text, height=800)
-
-
-
-
-
-# import streamlit as st
-# import pinecone
-# from langchain.embeddings.openai import OpenAIEmbeddings
-# from langchain.vectorstores import Pinecone, Chroma
-# from langchain.chains import RetrievalQA
-# from langchain.chat_models import ChatOpenAI
-# import tiktoken
-# import random
-
-# # Fetch the OpenAI API key from Streamlit secrets
-# openai_api_key = st.secrets["openai_api_key"]
-
-# # Fetch Pinecone API key and environment from Streamlit secrets
-# pinecone_api_key = st.secrets["pinecone_api_key"]
-# pinecone_environment = st.secrets["pinecone_environment"]
-
-# # Initialize Pinecone
-# pinecone.init(api_key=pinecone_api_key, environment=pinecone_environment)
-
-# # Define the name of the Pinecone index
-# index_name = 'mi-resource-qa'
-
-# # Initialize the OpenAI embeddings object with the hardcoded API key
-# embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
-
-# # Define functions
-# def insert_or_fetch_embeddings(index_name):
-#     if index_name in pinecone.list_indexes():
-#         vector_store = Pinecone.from_existing_index(index_name, embeddings)
-#         return vector_store
-#     else:
-#         raise ValueError(f"Index {index_name} does not exist. Please create it before fetching.")
-
-# # Initialize or fetch Pinecone vector store
-# vector_store = insert_or_fetch_embeddings(index_name)
-
-# # calculate embedding cost using tiktoken
-# def calculate_embedding_cost(text):
-#     import tiktoken
-#     enc = tiktoken.encoding_for_model('text-embedding-ada-002')
-#     total_tokens = len(enc.encode(text))
-#     # print(f'Total Tokens: {total_tokens}')
-#     # print(f'Embedding Cost in USD: {total_tokens / 1000 * 0.0004:.6f}')
-#     return total_tokens, total_tokens / 1000 * 0.0004
-
 
-# def ask_with_memory(vector_store, query, chat_history=[]):
-#     from langchain.chains import ConversationalRetrievalChain
-#     from langchain.chat_models import ChatOpenAI
+# CLEAR THE TEXT BOX
+with st.form("Question",clear_on_submit=True):
+    q = st.text_input(label='Ask a Question | Send a Prompt', placeholder=st.session_state.placeholder, value='', )
+    submitted = st.form_submit_button("Submit")
     
-#     llm = ChatOpenAI(model_name='gpt-3.5-turbo', temperature=1, openai_api_key=openai_api_key)
+    st.divider()
     
-#     # The retriever is created with metadata filter directly in search_kwargs
-#     # retriever = vector_store.as_retriever(search_type='similarity', search_kwargs={'k': 3, 'filter': {'source': {'$eq': 'https://mimtsstac.org/sites/default/files/session-documents/Intensifying%20Literacy%20Instruction%20-%20Essential%20Practices%20%28NATIONAL%29.pdf'}}})
-#     retriever = vector_store.as_retriever(search_type='similarity', search_kwargs={'k': 3, 'filter': {'source':'https://mimtsstac.org/sites/default/files/session-documents/Intensifying%20Literacy%20Instruction%20-%20Essential%20Practices%20%28NATIONAL%29.pdf'}})
-    
-#     chain= ConversationalRetrievalChain.from_llm(llm, retriever)
-#     result = chain({'question': query, 'chat_history': st.session_state['history']})
-#     # Append to chat history as a dictionary
-#     st.session_state['history'].append((query, result['answer']))
-    
-#     return (result['answer'])
-
-# # Initialize chat history
-# if 'history' not in st.session_state:
-#     st.session_state['history'] = []
-    
-# # # STREAMLIT APPLICATION SETUP WITH PASSWORD
-
-# # Define the correct password
-# # correct_password = "MiBLSi"
-
-# #Add the image with a specified width
-# image_width = 300  # Set the desired width in pixels
-# st.image('MTSS.ai_Logo.png', width=image_width)
-# st.subheader('Ink QA™ | Dynamic PDFs')
-
-# # Using Markdown for formatted text
-# st.markdown("""
-# Resource: **Intensifying Literacy Instruction: Essential Practices**
-# """, unsafe_allow_html=True)
-
-# with st.sidebar:
-#     # Password input field
-#     # password = st.text_input("Enter Password:", type="password")
-    
-#     st.image('mimtss.png', width=200)
-#     st.image('Literacy_Cover.png', width=200)
-#     st.link_button("View | Download", "https://mimtsstac.org/sites/default/files/session-documents/Intensifying%20Literacy%20Instruction%20-%20Essential%20Practices%20%28NATIONAL%29.pdf")
+    if submitted:
+        with st.spinner('Thinking...'):
+            answer = ask_with_memory(vector_store, q, st.session_state.history)
         
-#     Audio_Header_text = """
-#     **Tune into Dr. St. Martin's introduction**"""
-#     st.markdown(Audio_Header_text)
+        # st.write(q)
+        st.write(f"**{q}**")
         
-#     # Path or URL to the audio file
-#     audio_file_path = 'Audio_Introduction_Literacy.m4a'
-#     # Display the audio player widget
-#     st.audio(audio_file_path, format='audio/mp4', start_time=0)
+        import time
+        import random
+
+        def stream_answer():
+            for word in answer.split(" "):
+                yield word + " "
+                # time.sleep(0.02)
+                time.sleep(random.uniform(0.03, 0.08))
         
-#     # Citation text with Markdown formatting
-#     citation_Content_text = """
-#     **Citation**  
-#     St. Martin, K., Vaughn, S., Troia, G., Fien, & H., Coyne, M. (2023). *Intensifying literacy instruction: Essential practices, Version 2.0*. Lansing, MI: MiMTSS Technical Assistance Center, Michigan Department of Education.
+        st.write(stream_answer)
         
-#     **Table of Contents**  
-#     * **Introduction**: pg. 1  
-#     * **Intensifying Literacy Instruction: Essential Practices**: pg. 4  
-#     * **Purpose**: pg. 4  
-#     * **Practice 1**: Knowledge and Use of a Learning Progression for Developing Skilled Readers and Writers: pg. 6
-#     * **Practice 2**: Design and Use of an Intervention Platform as the Foundation for Effective Intervention: pg. 13
-#     * **Practice 3**: On-going Data-Based Decision Making for Providing and Intensifying Interventions: pg. 16
-#     * **Practice 4**: Adaptations to Increase the Instructional Intensity of the Intervention: pg. 20
-#     * **Practice 5**: Infrastructures to Support Students with Significant and Persistent Literacy Needs: pg. 24
-#     * **Motivation and Engagement**: pg. 28
-#     * **Considerations for Understanding How Students' Learning and Behavior are Enhanced**: pg. 28
-#     * **Summary**: pg. 29
-#     * **Endnotes**: pg. 30
-#     * **Acknowledgment**: pg. 39
-#     """
-#     st.markdown(citation_Content_text)
-
-# # if password == correct_password:
-# # Define a list of possible placeholder texts
-# placeholders = [
-#     'Example: Summarize the article in 200 words or less',
-#     'Example: What are the essential practices?',
-#     'Example: I am a teacher, why is this resource important?',
-#     'Example: How can this resource support my instruction in reading and writing?',
-#     'Example: Does this resource align with the learning progression for developing skilled readers and writers?',
-#     'Example: How does this resource address the needs of students scoring below the 20th percentile?',
-#     'Example: Are there assessment tools included in this resource to monitor student progress?',
-#     'Example: Does this resource provide guidance on data collection and analysis for monitoring student outcomes?',
-#     "Example: How can this resource be used to support students' social-emotional development?",
-#     "Example: How does this resource align with the district's literacy goals and objectives?",
-#     'Example: What research and evidence support the effectiveness of this resource?',
-#     'Example: Does this resource provide guidance on implementation fidelity'
-# ]
-
-# # Select a random placeholder from the list
-# if 'placeholder' not in st.session_state:
-#     st.session_state.placeholder = random.choice(placeholders)
-    
-# q = st.text_input(label='Ask a question or make a request ', value='', placeholder=st.session_state.placeholder)
-# # q = st.text_input(label='Ask a question or make a request ', value='')
-
-# k = 3  # Set k to 3
-
-# # # Initialize chat history if not present
-# # if 'history' not in st.session_state:
-# #     st.session_state.history = []
-    
-# if q:
-#     with st.spinner('Thinking...'):
-#         answer = ask_with_memory(vector_store, q, st.session_state.history)
-    
-#     # Display the response in a text area
-#     st.text_area('Response: ', value=answer, height=400, key="response_text_area")
-    
-#     st.success('Powered by MTSS GPT. AI can make mistakes. Consider checking important information.')
-
-#     # # Prepare chat history text for display
-#     # history_text = "\n\n".join(f"Q: {entry[0]}\nA: {entry[1]}" for entry in st.session_state.history)
-#     # Prepare chat history text for display in reverse order
-#     history_text = "\n\n".join(f"Q: {entry[0]}\nA: {entry[1]}" for entry in reversed(st.session_state.history))
-
-#     # Display chat history
-#     st.text_area('Chat History', value=history_text, height=800)
-
-
-
+        # Display the response in a text area
+        # st.text_area('Response: ', value=answer, height=400, key="response_text_area")
+        # OR to display as Markdown (interprets Markdown formatting)
+        # st.markdown(answer)
+            
+        st.success('Powered by MTSS GPT. AI can make mistakes. Consider checking important information.')
+        
+        st.divider()
 
+        # # Prepare chat history text for display
+        history_text = "\n\n".join(f"Q: {entry[0]}\nA: {entry[1]}" for entry in reversed(st.session_state.history))
 
+        # Display chat history
+        st.text_area('Chat History', value=history_text, height=800)