Update app.py

app.py

@@ -15,9 +15,22 @@ from keras.losses import MeanSquaredError, BinaryCrossentropy
 from keras.regularizers import l2, l1
 from keras.callbacks import Callback
 
-# Debug versions
-st.write(f"TensorFlow version: {tf.__version__}")
-st.write(f"Keras version: {tf.keras.__version__}")
+# Check TensorFlow and Keras versions with fallback
+try:
+    tf_version = tf.__version__
+    # Try multiple ways to get Keras version, accounting for TensorFlow integration
+    keras_version = None
+    if hasattr(tf.keras, '__version__'):
+        keras_version = tf.keras.__version__
+    elif hasattr(tf, 'keras') and hasattr(tf.keras, 'version'):
+        keras_version = tf.keras.version.__version__
+    else:
+        keras_version = "Keras version not available (bundled with TensorFlow)"
+    st.write(f"TensorFlow version: {tf_version}")
+    st.write(f"Keras version: {keras_version}")
+except AttributeError as e:
+    st.error(f"Error checking versions: {e}")
+    st.write("Falling back to default versions: TensorFlow ~2.15, Keras ~2.15")
 
 # Set wide layout and TensorFlow Playground CSS
 st.set_page_config(layout="wide")
@@ -111,7 +124,7 @@ with st.container():
     with col4:
         activation = st.selectbox("Activation", ["ReLU", "Sigmoid", "Tanh"], index=2)
     with col5:
-        batch_size = st.slider("Batch Size", 1, 30, 10)
+        batch_size = st.slider("Batch Size", 1, 10, 5)  # Reduced max batch size for Spaces
     
     # Row 2
     col6, col7, col8, col9, col10 = st.columns(5)
@@ -127,28 +140,28 @@ with st.container():
         st.button("Reset", key="reset_global", on_click=reset_session)
     st.markdown('</div>', unsafe_allow_html=True)
 
-# Dataset generation
-def generate_xor(n_samples):
+# Dataset generation (reduced sample size for performance)
+def generate_xor(n_samples=400):  # Reduced from 800 for performance
     X = np.random.rand(n_samples, 2) * 2 - 1
     y = np.logical_xor(X[:, 0] > 0, X[:, 1] > 0).astype(int)
     return X, y
 
-def generate_sine_wave(n_samples, noise):
+def generate_sine_wave(n_samples=400, noise):  # Reduced from 800
     X = np.linspace(-3, 3, n_samples).reshape(-1, 1)
     y = np.sin(X) + np.random.normal(0, noise / 100, X.shape)
     return np.hstack([X, X**2]), y.ravel()
 
 if problem_type == "Classification":
     if dataset_type == "Blobs":
-        fv, cv = make_blobs(n_samples=800, centers=2, n_features=2, cluster_std=1.5 + noise_level / 50, random_state=42)
+        fv, cv = make_blobs(n_samples=400, centers=2, n_features=2, cluster_std=1.5 + noise_level / 50, random_state=42)
     elif dataset_type == "Circles":
-        fv, cv = make_circles(n_samples=800, noise=noise_level / 250, factor=0.2)
+        fv, cv = make_circles(n_samples=400, noise=noise_level / 250, factor=0.2)
     elif dataset_type == "Spirals":
-        fv, cv = make_moons(n_samples=800, noise=noise_level / 250)
+        fv, cv = make_moons(n_samples=400, noise=noise_level / 250)
     elif dataset_type == "XOR":
-        fv, cv = generate_xor(800)
+        fv, cv = generate_xor(400)
 else:
-    fv, cv = generate_sine_wave(800, noise_level)
+    fv, cv = generate_sine_wave(400, noise_level)
 
 # Feature preprocessing
 std = StandardScaler()
@@ -362,7 +375,7 @@ with col_right:
             model = create_model(len(selected_features), st.session_state.hidden_layer_neurons)
             callback = OutputCallback(selected_data, cv)
             callback.model = model  # Explicitly set the model for the callback
-            model.fit(selected_data, cv, epochs=100,  # Reduced for Spaces
+            model.fit(selected_data, cv, epochs=50,  # Further reduced for Spaces
                       batch_size=batch_size, validation_split=1-train_ratio,
                       callbacks=[callback], verbose=0)
         except Exception as e: