fix: hmm 결과 출력 수정 (#52)

* remove: duplicated file

#44

* feat: spinner while HMM extract results
& refactor: unify model output format
#48 #44

* refactor: unify model output format

#44

streamlit/models/hmm.py

@@ -107,9 +107,9 @@ def visualize_hmm(self, predicted_close_prices):
             # ),
         ])   
         fig.update_layout(
-            title='Hidden Markov Model',
-            yaxis_title='Price (KRW)',
-            xaxis_title='Datetime',
+            title='Hidden Markov Model Prediction',
+            yaxis_title='Stock Price',
+            xaxis_title='Date',
             xaxis_rangeslider_visible=False,
             xaxis_type='category'
         )

streamlit/views/prediction.py

@@ -129,7 +129,7 @@ def get_stock_data(ticker, period, interval):
 
     st.plotly_chart(fig, use_container_width=True)
 
-    # Non-DL model matching 시각화
+    # AR model prediction 시각화
     def generate_stock_prediction(stock_ticker, period, interval):
     # Try to generate the predictions
         try:
@@ -181,8 +181,8 @@ def generate_stock_prediction(stock_ticker, period, interval):
     # Check if the data is not None
     if train_df is not None and (forecast >= 0).all() and (predictions >= 0).all():
         # Add a title to the stock prediction graph
-        st.markdown("## **Non DL Stock Prediction**")
-
+        layout = go.Layout(title='AR Prediction', xaxis=dict(title='Date'), yaxis=dict(title='Stock Price'))
+    
         # Create a plot for the stock prediction
         fig = go.Figure(
             data=[
@@ -214,7 +214,7 @@ def generate_stock_prediction(stock_ticker, period, interval):
                     mode="lines",
                     line=dict(color="green"),
                 ),
-            ]
+            ], layout=layout
         )
 
         # Customize the stock prediction graph
@@ -223,20 +223,16 @@ def generate_stock_prediction(stock_ticker, period, interval):
         # Use the native streamlit theme.
         st.plotly_chart(fig, use_container_width=True)
 
-    # If the data is None
-    else:
-        # Add a title to the stock prediction graph
-        st.markdown("## **Stock Prediction**")
-
-        # Add a message to the stock prediction graph
-        st.markdown("### **No data available for the selected stock**")    
-    if st.button("Start Non-DL model matching"):        
-        hmm = HMM(data)
-        predicted_close_prices = hmm.test_predictions()
-        fig = hmm.visualize_hmm(predicted_close_prices)
-        st.plotly_chart(fig, use_container_width=True) 
+    # HMM prediction 시각화
+    if st.button("Start HMM prediction"): 
+        with st.spinner('Wait for model output...'):    
+            hmm = HMM(data)
+            predicted_close_prices = hmm.test_predictions()
+            fig = hmm.visualize_hmm(predicted_close_prices)
+    st.plotly_chart(fig, use_container_width=True) 
 
-    # DL model matching 시각화
+
+    # DL model prediction 시각화
     data_close = data[["Close"]]
     data_close.fillna(method='pad')
 
@@ -273,12 +269,12 @@ def generate_stock_prediction(stock_ticker, period, interval):
     predicted_price_trace = go.Scatter(x=data_close[len(data_close)-len(y_test):].index, y=y_test_pred.flatten(), mode='lines', line=dict(color="green"), name='Predicted Stock Price')
     forecast_price_trace = go.Scatter(x=forecast_dates, y=forecast, mode='lines', line=dict(color="red"), name='Forecasted Stock Price')
 
-    layout = go.Layout(title='LSTM Stock Price Prediction', xaxis=dict(title='Date'), yaxis=dict(title='Stock Price'))
+    layout = go.Layout(title='LSTM Prediction', xaxis=dict(title='Date'), yaxis=dict(title='Stock Price'))
     fig2 = go.Figure(data=[real_price_trace, predicted_price_trace, forecast_price_trace], layout=layout)
 
     st.plotly_chart(fig2)
 
-    # Image-based CNN model matching 시각화
+    # Image-based CNN model prediction 시각화
     cnn_model_inference(company, ticker, period, interval)