Introduction to TensorFlow — Build Your First ML Model

In this tutorial, you'll learn about Introduction to TensorFlow. We cover key concepts, practical examples, and best practices to help you understand and apply this topic effectively.

What You'll Learn

Install TensorFlow, build a neural network with Keras, train it on the MNIST dataset, and evaluate its accuracy.

Why It Matters

TensorFlow is the most widely used ML framework in production. It powers Google Search, YouTube recommendations, and countless industrial ML systems.

Real-World Use

Image classification, text analysis, recommendation systems, and time series forecasting.

Installation

Pip install TensorFlow

Step 1: Load the Data

import TensorFlow as tf

# Load MNIST handwritten digits dataset
(x_train, y_train), (x_test, y_test) = tf.Keras.datasets.mnist.load_data()

# Normalize pixel values to 0-1
x_train, x_test = x_train / 255.0, x_test / 255.0

print(f"Training samples: {len(x_train)}")
print(f"Test samples: {len(x_test)}")
print(f"Image shape: {x_train[0].shape}")

Expected output:

Training samples: 60000
Test samples: 10000
Image shape: (28, 28)

Step 2: Build the Model

model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(input_shape=(28, 28)),
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dropout(0.2),
    tf.keras.layers.Dense(10, activation='softmax')
])

model.compile(
    optimizer='adam',
    loss='sparse_categorical_crossentropy',
    metrics=['accuracy']
)

Step 3: Train

history = model.fit(
    x_train, y_train,
    epochs=5,
    validation_data=(x_test, y_test)
)

Expected output (simplified):

Epoch 1/5   loss: 0.2933   accuracy: 0.9156   val_loss: 0.1324   val_accuracy: 0.9614
Epoch 2/5   loss: 0.1313   accuracy: 0.9614   val_loss: 0.0971   val_accuracy: 0.9706
...
Epoch 5/5   loss: 0.0687   accuracy: 0.9789   val_loss: 0.0746   val_accuracy: 0.9770

Step 4: Evaluate

test_loss, test_acc = model.evaluate(x_test, y_test, verbose=2)
print(f"\nTest accuracy: {test_acc:.4f}")

Expected output:

Test accuracy: 0.9770

97.7% accuracy on digits it's never seen before.

Step 5: Make a Prediction

import numpy as np

predictions = model.predict(x_test[:1])
predicted_digit = np.argmax(predictions[0])
print(f"Predicted digit: {predicted_digit}")