TensorFlow Basics

Tensors & Computation

At the core of TensorFlow are tensors: multi‑dimensional arrays similar to NumPy arrays but with automatic differentiation support.

import tensorflow as tf

x = tf.constant([[1., 2.], [3., 4.]])
y = tf.ones_like(x)

z = x + y
print(z.numpy())

Building Models with Keras Sequential

The Sequential API is the easiest way to stack layers and create feed‑forward neural networks.

from tensorflow.keras import Sequential
from tensorflow.keras.layers import Dense

model = Sequential([
    Dense(16, activation="relu", input_shape=(num_features,)),
    Dense(8, activation="relu"),
    Dense(1, activation="sigmoid")
])

model.compile(
    optimizer="adam",
    loss="binary_crossentropy",
    metrics=["accuracy"]
)

history = model.fit(X_train, y_train, epochs=20, batch_size=32,
                    validation_data=(X_val, y_val))

Practical Tips

• Normalize or standardize inputs for faster convergence.
• Use callbacks like EarlyStopping to prevent overfitting.
• Start with small models and increase complexity only if needed.

Next Steps with TensorFlow

• Explore convolutional neural networks (CNNs) for image data and recurrent / transformer models for sequence data.
• Use tf.data pipelines for efficient input pipelines on large datasets.
• Try saving and loading models with model.save() and tf.keras.models.load_model() for deployment.