Neural Networks & Deep LearningArchitecture & Training

import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers, callbacks
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.datasets import make_classification

# ── Reproducibility ───────────────────────────────────────
tf.random.set_seed(42)
np.random.seed(42)
print(f"TensorFlow Version: {tf.__version__}")

# ── Generate Synthetic Dataset ────────────────────────────
X, y = make_classification(
    n_samples=10000, n_features=20, n_informative=15,
    n_redundant=5, n_classes=2, random_state=42
)

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42, stratify=y
)
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test  = scaler.transform(X_test)

# ── Build the Model ───────────────────────────────────────
def build_model(input_dim, learning_rate=0.001):
    model = keras.Sequential([
        layers.Input(shape=(input_dim,)),
        layers.Dense(256, use_bias=False),
        layers.BatchNormalization(),
        layers.Activation("relu"),
        layers.Dropout(0.4),
        layers.Dense(128, use_bias=False),
        layers.BatchNormalization(),
        layers.Activation("relu"),
        layers.Dropout(0.3),
        layers.Dense(64, activation="relu"),
        layers.Dropout(0.2),
        layers.Dense(1, activation="sigmoid"),
    ], name="deep_classifier")

    model.compile(
        optimizer=keras.optimizers.Adam(learning_rate=learning_rate),
        loss="binary_crossentropy",
        metrics=["accuracy", keras.metrics.AUC(name="auc")]
    )
    return model

model = build_model(input_dim=X_train.shape[1])
model.summary()

# ── Define Callbacks ──────────────────────────────────────
cb_early_stop = callbacks.EarlyStopping(
    monitor="val_loss", patience=15,
    restore_best_weights=True, verbose=1
)
cb_reduce_lr = callbacks.ReduceLROnPlateau(
    monitor="val_loss", factor=0.5,
    patience=7, min_lr=1e-6, verbose=1
)
cb_checkpoint = callbacks.ModelCheckpoint(
    filepath="best_model.keras",
    monitor="val_auc", save_best_only=True,
    mode="max", verbose=1
)

# ── Train ─────────────────────────────────────────────────
history = model.fit(
    X_train, y_train,
    epochs=200, batch_size=256,
    validation_split=0.15,
    callbacks=[cb_early_stop, cb_reduce_lr, cb_checkpoint],
    verbose=1
)

# ── Evaluate ──────────────────────────────────────────────
test_loss, test_acc, test_auc = model.evaluate(X_test, y_test, verbose=0)
print(f"\n🏆 Test Accuracy : {test_acc:.4f}")
print(f"🏆 Test AUC      : {test_auc:.4f}")

# ── Plot Training Curves ──────────────────────────────────
fig, axes = plt.subplots(1, 3, figsize=(18, 5))
for ax, metric, title in zip(axes, ["loss","accuracy","auc"], ["Loss","Accuracy","AUC"]):
    ax.plot(history.history[metric],        label="Train", linewidth=2)
    ax.plot(history.history[f"val_{metric}"], label="Val", linewidth=2, linestyle="--")
    ax.set_title(f"Training {title}", fontsize=13)
    ax.set_xlabel("Epoch"); ax.legend(); ax.grid(alpha=0.3)
plt.tight_layout()
plt.savefig("training_curves.png", dpi=150)
plt.show()

from tensorflow.keras.datasets import cifar10

# ── Load and Normalize ────────────────────────────────────
(X_train_c, y_train_c), (X_test_c, y_test_c) = cifar10.load_data()
X_train_c = X_train_c.astype("float32") / 255.0
X_test_c  = X_test_c.astype("float32")  / 255.0

CLASS_NAMES = [
    "airplane","automobile","bird","cat","deer",
    "dog","frog","horse","ship","truck"
]

# ── Build CNN ─────────────────────────────────────────────
def build_cnn():
    model = keras.Sequential([
        layers.Input(shape=(32, 32, 3)),
        layers.Conv2D(32, (3,3), padding="same", activation="relu"),
        layers.BatchNormalization(),
        layers.Conv2D(32, (3,3), padding="same", activation="relu"),
        layers.MaxPooling2D((2,2)),
        layers.Dropout(0.25),

        layers.Conv2D(64, (3,3), padding="same", activation="relu"),
        layers.BatchNormalization(),
        layers.Conv2D(64, (3,3), padding="same", activation="relu"),
        layers.MaxPooling2D((2,2)),
        layers.Dropout(0.25),

        layers.Conv2D(128, (3,3), padding="same", activation="relu"),
        layers.BatchNormalization(),
        layers.GlobalAveragePooling2D(),

        layers.Dense(256, activation="relu"),
        layers.Dropout(0.5),
        layers.Dense(10, activation="softmax"),
    ], name="cifar10_cnn")

    model.compile(
        optimizer=keras.optimizers.Adam(1e-3),
        loss="sparse_categorical_crossentropy",
        metrics=["accuracy"]
    )
    return model

cnn = build_cnn()
cnn.summary()

# ── Data Augmentation ─────────────────────────────────────
data_aug = keras.Sequential([
    layers.RandomFlip("horizontal"),
    layers.RandomRotation(0.1),
    layers.RandomZoom(0.1),
    layers.RandomTranslation(0.1, 0.1),
], name="augmentation")

# ── Train ─────────────────────────────────────────────────
cnn_history = cnn.fit(
    data_aug(X_train_c), y_train_c,
    epochs=50, batch_size=128,
    validation_data=(X_test_c, y_test_c),
    callbacks=[
        callbacks.EarlyStopping(monitor="val_accuracy", patience=10, restore_best_weights=True),
        callbacks.ReduceLROnPlateau(monitor="val_loss", factor=0.5, patience=5)
    ], verbose=1
)
_, test_accuracy = cnn.evaluate(X_test_c, y_test_c, verbose=0)
print(f"\n🏆 CIFAR-10 Test Accuracy: {test_accuracy * 100:.2f}%")