Create mnist

mnist

@@ -0,0 +1,67 @@
+import tensorflow as tf
+from tensorflow.keras.losses import SparseCategoricalCrossentropy
+from tensorflow.keras.metrics import SparseCategoricalAccuracy
+from tensorflow.keras.optimizers import Adam
+from tensorflow.keras.datasets import mnist
+from KANtf import KAN
+
+# Load and preprocess data
+(x_train, y_train), (x_val, y_val) = mnist.load_data()
+x_train, x_val = (x_train / 255.0 - 0.5).astype('float32'), (x_val / 255.0 - 0.5).astype('float32')  # Normalize
+x_train, x_val = x_train.reshape(-1, 784), x_val.reshape(-1, 784)  # Flatten
+
+# Create TensorFlow datasets
+train_ds = tf.data.Dataset.from_tensor_slices((x_train, y_train)).shuffle(10000).batch(64)
+val_ds = tf.data.Dataset.from_tensor_slices((x_val, y_val)).batch(64)
+
+# Define the KAN model
+model = KAN([
+    {'in_features': 784, 'out_features': 64, 'grid_size': 5, 'spline_order': 3, 'activation': 'silu'},
+    {'in_features': 64, 'out_features': 10, 'grid_size': 5, 'spline_order': 3, 'activation': 'silu'}
+])
+
+# Compile the model with optimizer and loss function
+model.compile(optimizer=Adam(learning_rate=1e-3), loss=SparseCategoricalCrossentropy(from_logits=True))
+
+# Metrics
+train_loss = tf.keras.metrics.Mean(name='train_loss')
+val_loss = tf.keras.metrics.Mean(name='val_loss')
+val_accuracy = SparseCategoricalAccuracy(name='val_accuracy')
+
+# Lists to store metrics for plotting
+epoch_train_loss = []
+epoch_val_loss = []
+epoch_val_accuracy = []
+
+for epoch in range(epochs):
+    train_loss.reset_states()
+    val_loss.reset_states()
+    val_accuracy.reset_states()
+
+    # Existing training and validation loop here ...
+
+    # Append metrics after each epoch
+    epoch_train_loss.append(train_loss.result().numpy())
+    epoch_val_loss.append(val_loss.result().numpy())
+    epoch_val_accuracy.append(val_accuracy.result().numpy())
+
+    print(f'Epoch {epoch + 1}, Train Loss: {train_loss.result():.4f}, Validation Loss: {val_loss.result():.4f}, Validation Accuracy: {val_accuracy.result():.4f}')
+
+# After training, plot the metrics
+plt.figure(figsize=(8, 4))
+plt.subplot(1, 2, 1)
+plt.plot(epoch_train_loss, label='Train Loss')
+plt.plot(epoch_val_loss, label='Validation Loss')
+plt.title('Training and Validation Loss')
+plt.xlabel('Epoch')
+plt.ylabel('Loss')
+plt.legend()
+
+plt.subplot(1, 2, 2)
+plt.plot(epoch_val_accuracy, label='Validation Accuracy')
+plt.title('Validation Accuracy')
+plt.xlabel('Epoch')
+plt.ylabel('Accuracy')
+plt.legend()
+
+plt.show()