Update app.py

app.py

@@ -1,12 +1,10 @@
 import streamlit as st
 import pandas as pd
-import pickle
 from tensorflow.keras.models import Sequential
 from tensorflow.keras.layers import Dense, Dropout, Activation
 
-# Load the pre-trained model
-with open('model.pkl', 'rb') as file:
-    model = pickle.load(file)
+# Set up the Streamlit app
+st.title('Breast Cancer Prediction')
 
 # Default parameter values
 default_values = [17.99, 10.38, 122.8, 1001, 0.1184, 0.2776, 0.3001, 0.1471, 0.2419, 0.07871,
@@ -24,21 +22,11 @@ default_data = pd.DataFrame([default_values],
                                      'area_worst', 'smoothness_worst', 'compactness_worst', 'concavity_worst',
                                      'concave points_worst', 'symmetry_worst', 'fractal_dimension_worst'])
 
-# Set up the Streamlit app
-st.title('Breast Cancer Prediction')
-
 # Display the input form with default values
 st.subheader('Input Parameters')
 user_input = st.form(key='user_input_form')
 input_data = user_input.dataframe(default_data)
 
-# Make predictions when the 'Predict' button is clicked
-if user_input.form_submit_button('Predict'):
-    prediction = model.predict(input_data)
-    prediction_label = 'Malignant' if prediction[0] == 1 else 'Benign'
-    st.subheader('Prediction')
-    st.write(f'The lesion is predicted to be: {prediction_label}')
-
 # Implementing ANN
 ann_model = Sequential()
 ann_model.add(Dense(16, input_dim=30, activation='relu'))
@@ -48,7 +36,12 @@ ann_model.add(Dense(1, activation='sigmoid'))
 # Compiling the model
 ann_model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
 
-# Model summary
-st.subheader('Artificial Neural Network Model Summary')
-with st.echo():
-    ann_model.summary()
+# Load the saved model weights
+ann_model.load_weights('model_weights.h5')
+
+# Make predictions when the 'Predict' button is clicked
+if user_input.form_submit_button('Predict'):
+    prediction = ann_model.predict(input_data)
+    prediction_label = 'Malignant' if prediction[0] >= 0.5 else 'Benign'
+    st.subheader('Prediction')
+    st.write(f'The lesion is predicted to be: {prediction_label}')