Update pages/42_regression.py

pages/42_regression.py

@@ -1,49 +1,79 @@
-import streamlit as st
-import plotly.graph_objects as go
+import numpy as np
+import matplotlib.pyplot as plt
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from sklearn.model_selection import train_test_split
 
-# Title
-st.title("Interactive Equation Plotter")
+# Step 1: Create Synthetic Data
+np.random.seed(42)
+X = np.linspace(-10, 10, 1000)
+y = 2.5 * X + np.random.normal(0, 2, X.shape)  # Linear relation with noise
 
-# Sidebar sliders
-st.sidebar.header("Controls")
-w = st.sidebar.slider('W (slope)', min_value=-10.0, max_value=10.0, value=1.0, step=0.1)
-b = st.sidebar.slider('B (intercept)', min_value=-100.0, max_value=100.0, value=0.0, step=1.0)
+# Split into training and test sets
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
-# Initial x position
-x = st.slider('X position', min_value=-100.0, max_value=100.0, value=0.0, step=1.0)
+# Convert to PyTorch tensors
+X_train = torch.tensor(X_train, dtype=torch.float32).view(-1, 1)
+y_train = torch.tensor(y_train, dtype=torch.float32).view(-1, 1)
+X_test = torch.tensor(X_test, dtype=torch.float32).view(-1, 1)
+y_test = torch.tensor(y_test, dtype=torch.float32).view(-1, 1)
 
-# Calculate y based on the equation y = w * x + b
-y = w * x + b
+# Step 2: Define and Train a Neural Network Model
+class SimpleNN(nn.Module):
+    def __init__(self):
+        super(SimpleNN, self).__init__()
+        self.fc1 = nn.Linear(1, 10)
+        self.fc2 = nn.Linear(10, 1)
 
-# Create the plot
-fig = go.Figure()
+    def forward(self, x):
+        x = torch.relu(self.fc1(x))
+        x = self.fc2(x)
+        return x
 
-# Plot for the y-axis (top line)
-fig.add_shape(type="line", x0=-100, x1=100, y0=1, y1=1,
-              line=dict(color="blue", width=2, dash="dash"))
+# Initialize model, loss function, and optimizer
+model = SimpleNN()
+criterion = nn.MSELoss()
+optimizer = optim.Adam(model.parameters(), lr=0.01)
 
-# Plot for the x-axis (bottom line)
-fig.add_shape(type="line", x0=-100, x1=100, y0=-1, y1=-1,
-              line=dict(color="blue", width=2, dash="dash"))
+# Training loop
+epochs = 500
+losses = []
 
-# Plot the x point
-fig.add_trace(go.Scatter(x=[x], y=[-1], mode='markers', marker=dict(color='red', size=10), name="X point"))
+for epoch in range(epochs):
+    model.train()
+    optimizer.zero_grad()
+    outputs = model(X_train)
+    loss = criterion(outputs, y_train)
+    loss.backward()
+    optimizer.step()
+    losses.append(loss.item())
+    if (epoch + 1) % 50 == 0:
+        print(f'Epoch [{epoch+1}/{epochs}], Loss: {loss.item():.4f}')
 
-# Plot the y point
-fig.add_trace(go.Scatter(x=[y], y=[1], mode='markers', marker=dict(color='red', size=10), name="Y point"))
+# Step 3: Plot the Results
+# Plot the synthetic data and the model's predictions
+model.eval()
+with torch.no_grad():
+    predicted = model(X_test).numpy()
 
-# Update the layout
-fig.update_layout(
-    title=f'y = {w} * x + {b}',
-    xaxis=dict(range=[-100, 100]),
-    yaxis=dict(range=[-2, 2], showticklabels=False),
-    showlegend=False,
-    height=400,
-    margin=dict(t=50, b=10)
-)
+plt.figure(figsize=(12, 6))
 
-# Display the plot in Streamlit
-st.plotly_chart(fig)
+# Plot data and predictions
+plt.subplot(1, 2, 1)
+plt.scatter(X_test, y_test, label='Original data', alpha=0.5)
+plt.scatter(X_test, predicted, label='Fitted line', alpha=0.5)
+plt.title('Regression Results')
+plt.xlabel('X')
+plt.ylabel('y')
+plt.legend()
 
-# Add instruction to click and drag the point on x-axis
-st.write("Drag the red dot on the x-axis to change the x position.")
+# Plot training loss
+plt.subplot(1, 2, 2)
+plt.plot(losses)
+plt.title('Training Loss')
+plt.xlabel('Epoch')
+plt.ylabel('Loss')
+
+plt.tight_layout()
+plt.show()