Update app.py

app.py

@@ -1,6 +1,9 @@
 ### CSCI 4750/5750: regression models and gradient descent
 
 
+### CSCI 4750/5750: regression models
+
+
 import gradio as gr
 import matplotlib
 import matplotlib.pyplot as plt
@@ -14,7 +17,7 @@ def cal_mse(X,y,b,w):
   mse = np.mean((y_predict-y)**2)
   return mse
 
-def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, gradient_descent=False, learning_rate= 0.01, iteration=1000):
+def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, gradient_descent=False, learning_rate= 0.01, iteration=100):
   ### (1) generate simulated data points
   X = 2 * np.random.rand(100, 1)
   y = intercept + slope * X + np.random.randn(100, 1)
@@ -28,9 +31,9 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
   y_predict
 
   ### (4) Draw baseline linear Line
-  fig = plt.figure(figsize=(12,12))
+  fig = plt.figure(figsize=(12,18))
 
-  plt.subplot(2,1,1)
+  plt.subplot(3,1,1)
   plt.plot(X, y_predict, "r-", linewidth=2, label = "Line of best fit")
   plt.plot(X, y, "b.")
 
@@ -42,26 +45,27 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
     y_predict = X_new_b.dot(np.array([intercept_random, slope_random]))
     plt.plot(X_new, y_predict, "g-", linewidth=2, label = "Random line")
     
-
+    
     ### (4.3) Apply gradient desc
     if gradient_descent:
       b = intercept_random
       w = slope_random
 
       lr = learning_rate # learning rate
-      iteration = 1000
+      iteration = iteration
 
       # Store initial values for plotting.
       b_history = [b]
       w_history = [w]
 
+      train_mse = []
       # Iterations
       for i in range(iteration):
           b_grad = 0.0
           w_grad = 0.0
           for n in range(len(X)):        
-              b_grad = b_grad  - (y[n] - b - w*X[n])*1.0
-              w_grad = w_grad  - (y[n] - b - w*X[n])*X[n]
+              b_grad = b_grad  - (y[n,0] - b - w*X[n,0])*1.0
+              w_grad = w_grad  - (y[n,0] - b - w*X[n,0])*X[n,0]
           b_grad /= len(X)
           w_grad /= len(X)
           
@@ -72,6 +76,8 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
           # Store parameters for plotting
           b_history.append(b)
           w_history.append(w)
+          
+          train_mse.append(cal_mse(X,y,b,w))
   
   plt.xlabel("$x_1$", fontsize=22)
   plt.ylabel("$y$", rotation=0, fontsize=22)
@@ -86,7 +92,7 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
 
 
   ### (5) Visualize loss function
-  plt.subplot(2,1,2)
+  plt.subplot(3,1,2)
 
   ### (5.1) generate grid of parameters
   b = np.arange(-10,10,0.1) #bias
@@ -132,6 +138,19 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
   plt.yticks(fontsize=18)
   plt.xlim(-10,10)
   plt.ylim(-10,10)
+
+
+  ### 6. Visualize the learning curves 
+  if gradient_descent:
+      plt.subplot(3,1,3)
+      plt.plot(train_mse,label="train_loss (lr="+str(learning_rate)+")")
+      plt.xlabel('Iteration',fontweight="bold",fontsize = 22)
+      plt.ylabel('Loss',fontweight="bold",fontsize = 22)
+      plt.title("Learning curve: Loss VS Epochs",fontweight="bold",fontsize = 22)
+      plt.legend(fontsize=18)
+      plt.xticks(fontsize=18)
+      plt.yticks(fontsize=18)
+
   #plt.show()
   fig.tight_layout()
   plt.savefig('plot_line.png', dpi=300)
@@ -139,6 +158,9 @@ def gradient_descent(intercept=4, slope=3, intercept_random=4, slope_random=3, g
 
 
 
+
+
+
 #### Define input component
 input_intercept = gr.inputs.Slider(1, 8, step=0.5, default=4, label='(Baseline) Intercept')
 input_slope = gr.inputs.Slider(-8, 8, step=0.5, default=2.8, label='(Baseline) Slope')