changes to Data Set

app.py

@@ -8,6 +8,7 @@ from keras.models import Sequential
 import matplotlib.pyplot as plt
 from keras.layers import Dense
 import streamlit as st
+import numpy as np
 import io
 
 
@@ -19,7 +20,7 @@ def model_MLP(X_train,y_train,X_test,layers, nodes, activation, solver, rate, it
     model = Sequential() 
 
     # Adding first layers.
-    model.add(Dense(nodes, activation=activation, input_dim=X_train.shape[1]))
+    model.add(Dense(nodes, activation=activation, input_dim=1))
 
     # Adding remaining hidden layers.
     for i in range(layers-1):
@@ -82,7 +83,7 @@ if __name__ == '__main__':
     with right_column:
 
         # slider for adding noise.
-        noise = st.slider('Noise', min_value=0,max_value= 100,value=50,step=10)
+        noise = st.slider('Noise', min_value=0,max_value= 100,value=20,step=10)
         # slider for test-train split.                     
         split = st.slider('Test-Train Split', min_value=0.1,max_value= 0.9,value=0.3,step=0.1) 
         # selectbox for solver/optimizer.     
@@ -91,12 +92,15 @@ if __name__ == '__main__':
         rate = float(st.selectbox('Learning Rate',('0.001','0.003','0.01','0.03','0.1','0.3','1.0'),index=3))   
 
     # Generating regression data.
-    X, y = make_regression(n_samples=500, n_features=1, noise=noise,random_state=42,bias=3)
+    # X, y = make_regression(n_samples=100, n_features=1, noise=noise,random_state=42,bias=3)
+    X=np.linspace(0,50,100)
+    y = np.sin(X) + X + X*np.random.normal(0,1,100)/5
 
     # Split data into training and test sets.    
     X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=split,random_state=42)
 
     # Predicting the test data.
+    X_test.sort(axis=0)
     y_hat,model = model_MLP(X_train,y_train,X_test,layers, nodes, activation, solver, rate, iter)
 
     # Printing Model Architecture.
@@ -130,6 +134,7 @@ if __name__ == '__main__':
             
             # write the graph to the app.
             st.pyplot(fig1)
+            plt.savefig('plot_1.jpg')
 
         with right_graph:
 
@@ -137,8 +142,9 @@ if __name__ == '__main__':
             st.write('Test Data set')
 
             fig2, ax2 = plt.subplots(1)
-            ax2.scatter(X_test, y_test, label='test',color='blue',alpha=0.4)
+            ax2.scatter(X_test, y_test, label='test',color='blue',alpha=0.6)
             ax2.scatter(X_test, y_hat, label='prediction',c='red',alpha=0.6,edgecolors='black')
+            ax2.plot(X_test, y_hat, label='prediction',c='red',alpha=0.6)
 
             # setting the labels and title of the graph.
             ax2.set_xlabel('X')
@@ -148,7 +154,7 @@ if __name__ == '__main__':
             
             # write the graph to the app.
             st.pyplot(fig2)
-        
+            plt.savefig('plot_2.jpg')
 
         # Printing the Errors.
         st.subheader('Errors')

requirements.txt

@@ -2,4 +2,5 @@ scikit-learn==1.2.0
 keras==2.12.0
 streamlit==1.19.0
 tensorflow==2.12.0
-matplotlib==3.6.3
+matplotlib==3.6.3
+numpy==1.23.5