button added

app.py

@@ -59,6 +59,7 @@ special_threshold = st.sidebar.number_input(
 st.sidebar.success(
     "The 'distances' score indicates the proximity of your question to our database questions (lower is better). The 'ai_judge' ranks the similarity between user's question and database answers independently (higher is better)."
 )
+submit_button = st.sidebar.button("Submit", type="primary")
 clear_button = st.sidebar.button("Clear Conversation", key="clear")
 if clear_button:
     st.session_state.messages = []
@@ -114,99 +115,100 @@ with st.spinner("Loading, please be patient with us ... 🙏"):
 
 
 # React to user input
-if prompt := st.chat_input(initial_input):
-    with st.spinner("Loading, please be patient with us ... 🙏"):
-        # Display user message in chat message container
-        st.chat_message("user").markdown(prompt)
-        # Add user message to chat history
-        st.session_state.messages.append({"role": "user", "content": prompt})
-
-        question = prompt
-        begin_t = time.time()
-        results = collection.query(query_texts=question, n_results=5)
-        end_t = time.time()
-        st.success(f"Query answser. | Time: {end_t - begin_t} sec")
-        idx = results["ids"][0]
-        idx = [int(i) for i in idx]
-        ref = pd.DataFrame(
-            {
-                "idx": idx,
-                "questions": [dataset["train"]["questions"][i] for i in idx],
-                "answers": [dataset["train"]["answers"][i] for i in idx],
-                "distances": results["distances"][0],
-            }
-        )
-        # special_threshold = st.sidebar.slider('How old are you?', 0, 0.6, 0.1) # 0.3
-        filtered_ref = ref[ref["distances"] < special_threshold]
-        if filtered_ref.shape[0] > 0:
-            st.success("There are highly relevant information in our database.")
-            ref_from_db_search = filtered_ref["answers"].str.cat(sep=" ")
-            final_ref = filtered_ref
-        else:
-            st.warning(
-                "The database may not have relevant information to help your question so please be aware of hallucinations."
+if submit_button:
+    if prompt := st.chat_input(initial_input):
+        with st.spinner("Loading, please be patient with us ... 🙏"):
+            # Display user message in chat message container
+            st.chat_message("user").markdown(prompt)
+            # Add user message to chat history
+            st.session_state.messages.append({"role": "user", "content": prompt})
+
+            question = prompt
+            begin_t = time.time()
+            results = collection.query(query_texts=question, n_results=5)
+            end_t = time.time()
+            st.success(f"Query answser. | Time: {end_t - begin_t} sec")
+            idx = results["ids"][0]
+            idx = [int(i) for i in idx]
+            ref = pd.DataFrame(
+                {
+                    "idx": idx,
+                    "questions": [dataset["train"]["questions"][i] for i in idx],
+                    "answers": [dataset["train"]["answers"][i] for i in idx],
+                    "distances": results["distances"][0],
+                }
             )
-            ref_from_db_search = ref["answers"].str.cat(sep=" ")
-            final_ref = ref
-
-        if option == "YSA":
-            try:
+            # special_threshold = st.sidebar.slider('How old are you?', 0, 0.6, 0.1) # 0.3
+            filtered_ref = ref[ref["distances"] < special_threshold]
+            if filtered_ref.shape[0] > 0:
+                st.success("There are highly relevant information in our database.")
+                ref_from_db_search = filtered_ref["answers"].str.cat(sep=" ")
+                final_ref = filtered_ref
+            else:
+                st.warning(
+                    "The database may not have relevant information to help your question so please be aware of hallucinations."
+                )
+                ref_from_db_search = ref["answers"].str.cat(sep=" ")
+                final_ref = ref
+
+            if option == "YSA":
+                try:
+                    begin_t = time.time()
+                    llm_response = llama2_7b_ysa(question)
+                    end_t = time.time()
+                    st.success(f"Running LLM. | Time: {end_t - begin_t} sec")
+                except:
+                    st.warning("Sorry, the inference endpoint is temporarily down. 😔")
+                    llm_response = "NA."
+            else:
+                st.warning(
+                    "Apologies! We are in the progress of fine-tune the model, so it's currently unavailable. ⚙️"
+                )
+                llm_response = "NA"
+
+            finetuned_llm_guess = ["from_llm", question, llm_response, 0]
+            final_ref.loc[-1] = finetuned_llm_guess
+            final_ref = final_ref.reset_index()
+
+            # add ai judge as additional rating
+            if run_ai_judge == "Yes":
+                independent_ai_judge_score = []
                 begin_t = time.time()
-                llm_response = llama2_7b_ysa(question)
+                for i in range(final_ref.shape[0]):
+                    this_content = final_ref["answers"][i]
+                    if len(this_content) > 3:
+                        arr1 = openai_text_embedding(question)
+                        arr2 = openai_text_embedding(this_content)
+                        # this_score = calculate_sts_openai_score(question, this_content)
+                        this_score = quantized_influence(arr1, arr2)
+                    else:
+                        this_score = 0
+                    independent_ai_judge_score.append(this_score)
+
+                final_ref["ai_judge"] = independent_ai_judge_score
+
                 end_t = time.time()
-                st.success(f"Running LLM. | Time: {end_t - begin_t} sec")
-            except:
-                st.warning("Sorry, the inference endpoint is temporarily down. 😔")
-                llm_response = "NA."
-        else:
-            st.warning(
-                "Apologies! We are in the progress of fine-tune the model, so it's currently unavailable. ⚙️"
-            )
-            llm_response = "NA"
+                st.success(f"Using AI Judge. | Time: {end_t - begin_t} sec")
 
-        finetuned_llm_guess = ["from_llm", question, llm_response, 0]
-        final_ref.loc[-1] = finetuned_llm_guess
-        final_ref = final_ref.reset_index()
+            engineered_prompt = f"""
+                Based on the context: {ref_from_db_search}
 
-        # add ai judge as additional rating
-        if run_ai_judge == "Yes":
-            independent_ai_judge_score = []
-            begin_t = time.time()
-            for i in range(final_ref.shape[0]):
-                this_content = final_ref["answers"][i]
-                if len(this_content) > 3:
-                    arr1 = openai_text_embedding(question)
-                    arr2 = openai_text_embedding(this_content)
-                    # this_score = calculate_sts_openai_score(question, this_content)
-                    this_score = quantized_influence(arr1, arr2)
-                else:
-                    this_score = 0
-                independent_ai_judge_score.append(this_score)
-
-            final_ref["ai_judge"] = independent_ai_judge_score
+                answer the user question: {question}
 
+                Answer the question directly (don't say "based on the context, ...")
+            """
+
+            begin_t = time.time()
+            answer = call_chatgpt(engineered_prompt)
             end_t = time.time()
-            st.success(f"Using AI Judge. | Time: {end_t - begin_t} sec")
-
-        engineered_prompt = f"""
-            Based on the context: {ref_from_db_search}
-
-            answer the user question: {question}
-
-            Answer the question directly (don't say "based on the context, ...")
-        """
-
-        begin_t = time.time()
-        answer = call_chatgpt(engineered_prompt)
-        end_t = time.time()
-        st.success(f"Final API Call. | Time: {end_t - begin_t} sec")
-        response = answer
-
-    # Display assistant response in chat message container
-    with st.chat_message("assistant"):
-        with st.spinner("Wait for it..."):
-            st.markdown(response)
-            with st.expander("See reference:"):
-                st.table(final_ref)
-    # Add assistant response to chat history
-    st.session_state.messages.append({"role": "assistant", "content": response})
+            st.success(f"Final API Call. | Time: {end_t - begin_t} sec")
+            response = answer
+
+        # Display assistant response in chat message container
+        with st.chat_message("assistant"):
+            with st.spinner("Wait for it..."):
+                st.markdown(response)
+                with st.expander("See reference:"):
+                    st.table(final_ref)
+        # Add assistant response to chat history
+        st.session_state.messages.append({"role": "assistant", "content": response})