Update app.py

app.py

@@ -1,15 +1,20 @@
 import gradio as gr
 import numpy as np
 import plotly.graph_objects as go
-import matplotlib.pyplot as plt
 from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier
 from sklearn.metrics import mean_squared_error, accuracy_score
-from sklearn.tree import plot_tree
 
 # ------------------------------------------------
 # DATA GENERATORS
 # ------------------------------------------------
 
+def generate_2d_regression(n_points, noise):
+    n_points = int(n_points)
+    X = np.linspace(0, 10, n_points)
+    y = 2.5 * X + 5 + np.random.randn(n_points) * noise
+    return X.reshape(-1, 1), y
+
+
 def generate_3d_regression(n_points, noise):
     n_points = int(n_points)
     x1 = np.linspace(0, 10, n_points)
@@ -19,7 +24,7 @@ def generate_3d_regression(n_points, noise):
 
     X_flat = np.column_stack((X1.ravel(), X2.ravel()))
     Z_flat = Z.ravel()
-    return X1, X2, X_flat, Z_flat
+    return X, X1, X2, X_flat, Z_flat
 
 
 def generate_classification(n_points, noise):
@@ -31,54 +36,87 @@ def generate_classification(n_points, noise):
 
 
 # ------------------------------------------------
-# INTERACTIVE 3D RANDOM FOREST
+# 2D RANDOM FOREST REGRESSION
 # ------------------------------------------------
 
-def interactive_3d(n_points, noise, n_estimators, max_depth):
+def rf_2d_view(n_points, noise, n_estimators, max_depth):
     n_estimators = int(n_estimators)
     max_depth = int(max_depth)
 
-    X1, X2, X_flat, Z_flat = generate_3d_regression(n_points, noise)
+    X, y = generate_2d_regression(n_points, noise)
 
     rf = RandomForestRegressor(n_estimators=n_estimators, max_depth=max_depth, random_state=42)
-    rf.fit(X_flat, Z_flat)
+    rf.fit(X, y)
 
-    Z_pred = rf.predict(X_flat).reshape(X1.shape)
-    mse = mean_squared_error(Z_flat, Z_pred.ravel())
+    y_pred = rf.predict(X)
+    mse = mean_squared_error(y, y_pred)
 
     fig = go.Figure()
 
-    fig.add_surface(x=X1, y=X2, z=Z_pred, colorscale="Blues", opacity=0.9)
+    fig.add_scatter(x=X.flatten(), y=y, mode="markers", name="Data")
+    fig.add_scatter(x=X.flatten(), y=y_pred, mode="lines", name="RF Prediction")
 
     fig.update_layout(
-        title="Interactive 3D Random Forest Surface",
-        scene=dict(bgcolor="#0b1e3d"),
+        title="2D Random Forest Regression",
         paper_bgcolor="#0b1e3d",
+        plot_bgcolor="#0b1e3d",
         font=dict(color="white")
     )
 
-    return fig, f"MSE: {mse:.4f}", rf
+    return fig, f"MSE: {mse:.4f}"
 
 
 # ------------------------------------------------
-# SINGLE TREE VISUALIZATION
+# 3D RANDOM FOREST REGRESSION (ROTATING CAMERA)
 # ------------------------------------------------
 
-def show_tree(rf_model):
-    if rf_model is None:
-        return None
+def rf_3d_view(n_points, noise, n_estimators, max_depth):
+    n_estimators = int(n_estimators)
+    max_depth = int(max_depth)
+
+    _, X1, X2, X_flat, Z_flat = generate_3d_regression(n_points, noise)
+
+    rf = RandomForestRegressor(n_estimators=n_estimators, max_depth=max_depth, random_state=42)
+    rf.fit(X_flat, Z_flat)
 
-    tree = rf_model.estimators_[0]
+    Z_pred = rf.predict(X_flat).reshape(X1.shape)
+    mse = mean_squared_error(Z_flat, Z_pred.ravel())
 
-    fig, ax = plt.subplots(figsize=(10, 6))
-    plot_tree(tree, ax=ax, filled=True)
-    ax.set_title("Single Decision Tree from Random Forest")
+    fig = go.Figure()
 
-    return fig
+    fig.add_surface(x=X1, y=X2, z=Z_pred, colorscale="Blues", opacity=0.9)
+
+    # Smooth rotating camera animation
+    frames = []
+    for angle in range(0, 360, 10):
+        frames.append(go.Frame(layout=dict(scene_camera=dict(eye=dict(
+            x=np.cos(np.radians(angle)) * 2,
+            y=np.sin(np.radians(angle)) * 2,
+            z=1.2
+        )))))
+
+    fig.frames = frames
+
+    fig.update_layout(
+        title="3D Random Forest Regression",
+        scene=dict(bgcolor="#0b1e3d"),
+        paper_bgcolor="#0b1e3d",
+        font=dict(color="white"),
+        updatemenus=[dict(
+            type="buttons",
+            showactive=False,
+            buttons=[dict(label="Rotate 3D",
+                          method="animate",
+                          args=[None, {"frame": {"duration": 60, "redraw": True},
+                                       "fromcurrent": True}])]
+        )]
+    )
+
+    return fig, f"MSE: {mse:.4f}"
 
 
 # ------------------------------------------------
-# CLASSIFICATION VIEW
+# CLASSIFICATION VIEW (CLEAR DECISION REGIONS)
 # ------------------------------------------------
 
 def classification_view(n_points, noise, n_estimators, max_depth):
@@ -90,20 +128,24 @@ def classification_view(n_points, noise, n_estimators, max_depth):
     rf = RandomForestClassifier(n_estimators=n_estimators, max_depth=max_depth, random_state=42)
     rf.fit(X, y)
 
-    y_pred = rf.predict(X)
-    acc = accuracy_score(y, y_pred)
+    # Mesh grid for decision boundary
+    xx, yy = np.meshgrid(
+        np.linspace(X[:, 0].min() - 1, X[:, 0].max() + 1, 100),
+        np.linspace(X[:, 1].min() - 1, X[:, 1].max() + 1, 100)
+    )
+
+    Z = rf.predict(np.c_[xx.ravel(), yy.ravel()]).reshape(xx.shape)
+
+    acc = accuracy_score(y, rf.predict(X))
 
     fig = go.Figure()
 
-    fig.add_trace(go.Scatter(
-        x=X[:, 0],
-        y=X[:, 1],
-        mode="markers",
-        marker=dict(color=y_pred, colorscale="Blues"),
-    ))
+    fig.add_contour(x=xx[0], y=yy[:, 0], z=Z, showscale=False, opacity=0.4, colorscale="Blues")
+
+    fig.add_scatter(x=X[:, 0], y=X[:, 1], mode="markers", marker=dict(color=y, colorscale="Blues"), name="Data")
 
     fig.update_layout(
-        title="Random Forest Classification",
+        title="Random Forest Classification Boundary",
         paper_bgcolor="#0b1e3d",
         plot_bgcolor="#0b1e3d",
         font=dict(color="white")
@@ -113,41 +155,44 @@ def classification_view(n_points, noise, n_estimators, max_depth):
 
 
 # ------------------------------------------------
-# GRADIO UI (AUTO RUN, BLUE THEME)
+# GRADIO UI (AUTO-RUN, BEGINNER FRIENDLY)
 # ------------------------------------------------
 
 with gr.Blocks() as demo:
 
-    gr.Markdown("# 🌲 Random Forest Teaching Dashboard")
-
-    with gr.Tab("🌐 3D Regression"):
-        n_points = gr.Slider(20, 100, 40, step=1, label="Points")
-        noise = gr.Slider(0.0, 3.0, 1.0, label="Noise")
-        n_estimators = gr.Slider(10, 100, 50, step=1, label="Trees")
-        max_depth = gr.Slider(2, 15, 8, step=1, label="Depth")
-
-        plot3d = gr.Plot()
-        mse_text = gr.Markdown()
-        tree_plot = gr.Plot()
+    gr.Markdown("# 🌲 Random Forest Learning Dashboard (2D • 3D • Classification)")
 
-        def run_3d(n_points, noise, n_estimators, max_depth):
-            fig, text, rf = interactive_3d(n_points, noise, n_estimators, max_depth)
-            tree_fig = show_tree(rf)
-            return fig, text, tree_fig
+    n_points = gr.Slider(20, 100, 40, step=1, label="Number of Points")
+    noise = gr.Slider(0.0, 3.0, 1.0, label="Noise Level")
+    n_estimators = gr.Slider(10, 100, 50, step=1, label="Number of Trees")
+    max_depth = gr.Slider(2, 15, 8, step=1, label="Tree Depth")
 
-        for inp in [n_points, noise, n_estimators, max_depth]:
-            inp.change(run_3d, [n_points, noise, n_estimators, max_depth], [plot3d, mse_text, tree_plot])
+    with gr.Tab("📈 2D Regression"):
+        plot2d = gr.Plot()
+        mse2d = gr.Markdown()
 
-        demo.load(run_3d, [n_points, noise, n_estimators, max_depth], [plot3d, mse_text, tree_plot])
+    with gr.Tab("🌐 3D Regression"):
+        plot3d = gr.Plot()
+        mse3d = gr.Markdown()
 
     with gr.Tab("🧩 Classification"):
-        cls_plot = gr.Plot()
-        cls_text = gr.Markdown()
+        plot_cls = gr.Plot()
+        acc_cls = gr.Markdown()
+
+    def run_all(n_points, noise, n_estimators, max_depth):
+        fig2d, mse_text = rf_2d_view(n_points, noise, n_estimators, max_depth)
+        fig3d, mse3d_text = rf_3d_view(n_points, noise, n_estimators, max_depth)
+        fig_cls, acc_text = classification_view(n_points, noise, n_estimators, max_depth)
+        return fig2d, mse_text, fig3d, mse3d_text, fig_cls, acc_text
 
-        for inp in [n_points, noise, n_estimators, max_depth]:
-            inp.change(classification_view, [n_points, noise, n_estimators, max_depth], [cls_plot, cls_text])
+    for inp in [n_points, noise, n_estimators, max_depth]:
+        inp.change(run_all,
+                   [n_points, noise, n_estimators, max_depth],
+                   [plot2d, mse2d, plot3d, mse3d, plot_cls, acc_cls])
 
-        demo.load(classification_view, [n_points, noise, n_estimators, max_depth], [cls_plot, cls_text])
+    demo.load(run_all,
+              [n_points, noise, n_estimators, max_depth],
+              [plot2d, mse2d, plot3d, mse3d, plot_cls, acc_cls])
 
 
-demo.launch(theme=gr.themes.Soft(primary_hue="blue"))
+demo.launch(theme=gr.themes.Soft(primary_hue="blue"))