Update app.py

app.py

@@ -5,6 +5,7 @@ import numpy as np
 import yfinance as yf
 import plotly.express as px
 import plotly.graph_objects as go
+import os
 
 from sklearn.preprocessing import MinMaxScaler
 from tensorflow.keras.models import Sequential
@@ -21,11 +22,16 @@ end_date = st.sidebar.date_input('End Date')
 # --- MAIN PAGE
 st.header('Cryptocurrency Prediction')
 
-col1, col2, = st.columns([1,9])
+col1, col2 = st.columns([1, 9])
 with col1:
-  st.image('icons/'+ ticker +'.png', width=75)
+    image_path = 'icons/' + ticker + '.png'
+    if os.path.exists(image_path):
+        st.image(image_path, width=75)
+    else:
+        st.warning(f"Image for {ticker} not found.")
+
 with col2:
-  st.write(f" ## {ticker}")
+    st.write(f" ## {ticker}")
 
 ticker_obj = yf.Ticker(ticker)
 
@@ -63,7 +69,7 @@ split_ratio = 0.8  # Ratio of training set to total data
 zero_base = True
 lstm_neurons = 50
 epochs = 100
-batch_size = 128 #32
+batch_size = 128
 loss = 'mean_squared_error'
 dropout = 0.24
 optimizer = 'adam'
@@ -91,8 +97,12 @@ model = build_lstm_model(X_train, output_size=1, neurons=lstm_neurons, dropout=d
 # Saved Weights
 file_path = "./LSTM_" + ticker + "_weights.h5"
 
-# Loads the weights
-model.load_weights(file_path)
+# Check if weights file exists
+if os.path.exists(file_path):
+    # Loads the weights
+    model.load_weights(file_path)
+else:
+    st.warning(f"Weights file for {ticker} not found.")
 
 # Step 4: Make predictions
 preds = model.predict(X_test)
@@ -110,13 +120,13 @@ y_test = scaler.inverse_transform(y_test)
 
 fig = px.line(x=model_data.index[-len(y_test):],
               y=[y_test.flatten(), preds.flatten()])
-newnames = {'wide_variable_0':'Real Values', 'wide_variable_1': 'Predictions'}
-fig.for_each_trace(lambda t: t.update(name = newnames[t.name],
-                                      legendgroup = newnames[t.name],
-                                      hovertemplate = t.hovertemplate.replace(t.name, newnames[t.name])))
+newnames = {'wide_variable_0': 'Real Values', 'wide_variable_1': 'Predictions'}
+fig.for_each_trace(lambda t: t.update(name=newnames[t.name],
+                                      legendgroup=newnames[t.name],
+                                      hovertemplate=t.hovertemplate.replace(t.name, newnames[t.name])))
 fig.update_layout(
     xaxis_title="Date",
-    yaxis_title=ticker+" Price",
+    yaxis_title=ticker + " Price",
     legend_title=" ")
 st.write(fig)
 
@@ -125,34 +135,34 @@ st.write(fig)
 about_data, news = st.tabs(["About", "News"])
 
 with about_data:
-  # Candlestick
-  raw_data = ticker_obj.history(start=start_date, end=end_date)
-  fig = go.Figure(data=[go.Candlestick(x=raw_data.index,
-                  open=raw_data['Open'],
-                  high=raw_data['High'],
-                  low=raw_data['Low'],
-                  close=raw_data['Close'])])
-  fig.update_layout(
-                  title=ticker + " candlestick : Open, High, Low and Close",
-                  yaxis_title=ticker + ' Price')
-  st.plotly_chart(fig)
-
-  # Table
-  history_data = raw_data.copy()
-
-  # Formating index Date
-  history_data.index = pd.to_datetime(history_data.index, format='%Y-%m-%d %H:%M:%S').date
-  history_data.index.name = "Date"
-  history_data.sort_values(by='Date', ascending=False, inplace=True)
-  st.write(history_data)
+    # Candlestick
+    raw_data = ticker_obj.history(start=start_date, end=end_date)
+    fig = go.Figure(data=[go.Candlestick(x=raw_data.index,
+                                         open=raw_data['Open'],
+                                         high=raw_data['High'],
+                                         low=raw_data['Low'],
+                                         close=raw_data['Close'])])
+    fig.update_layout(
+        title=ticker + " candlestick : Open, High, Low and Close",
+        yaxis_title=ticker + ' Price')
+    st.plotly_chart(fig)
+
+    # Table
+    history_data = raw_data.copy()
+
+    # Formating index Date
+    history_data.index = pd.to_datetime(history_data.index, format='%Y-%m-%d %H:%M:%S').date
+    history_data.index.name = "Date"
+    history_data.sort_values(by='Date', ascending=False, inplace=True)
+    st.write(history_data)
 
 with news:
-  sNews = StockNews(ticker, save_news=False)
-  sNews_df = sNews.read_rss()
-
-  # Showing most recent news
-  for i in range(10):
-    st.subheader(f"{i+1} - {sNews_df['title'][i]}")
-    st.write(sNews_df['summary'][i])
-    date_object = datetime.strptime(sNews_df['published'][i], '%a, %d %b %Y %H:%M:%S %z')
-    st.write(f"_{date_object.strftime('%A')}, {date_object.date()}_")
+    sNews = StockNews(ticker, save_news=False)
+    sNews_df = sNews.read_rss()
+
+    # Showing most recent news
+    for i in range(10):
+        st.subheader(f"{i+1} - {sNews_df['title'][i]}")
+        st.write(sNews_df['summary'][i])
+        date_object = datetime.strptime(sNews_df['published'][i], '%a, %d %b %Y %H:%M:%S %z')
+        st.write(f"_{date_object.strftime('%A')}, {date_object.date()}_")