TensorFlow

End-to-end machine learning platform with Keras integration.

When to Use

• Production ML pipelines
• Model deployment (TensorFlow Serving)
• Mobile ML (TensorFlow Lite)
• Large-scale training

Quick Start

import tensorflow as tf
from tensorflow import keras

model = keras.Sequential([
    keras.layers.Dense(128, activation='relu', input_shape=(784,)),
    keras.layers.Dense(10, activation='softmax')
])

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

Core Concepts

Keras Functional API

from tensorflow import keras
from tensorflow.keras import layers

inputs = keras.Input(shape=(784,))
x = layers.Dense(256, activation='relu')(inputs)
x = layers.Dropout(0.2)(x)
x = layers.Dense(128, activation='relu')(x)
outputs = layers.Dense(10, activation='softmax')(x)

model = keras.Model(inputs, outputs, name='classifier')

Custom Training

@tf.function
def train_step(x, y):
    with tf.GradientTape() as tape:
        predictions = model(x, training=True)
        loss = loss_fn(y, predictions)

    gradients = tape.gradient(loss, model.trainable_variables)
    optimizer.apply_gradients(zip(gradients, model.trainable_variables))
    return loss

for epoch in range(epochs):
    for x_batch, y_batch in dataset:
        loss = train_step(x_batch, y_batch)

Common Patterns

Data Pipeline

dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train))
dataset = dataset.shuffle(buffer_size=1024)
dataset = dataset.batch(32)
dataset = dataset.prefetch(tf.data.AUTOTUNE)

Model Saving

# SavedModel format
model.save('saved_model/my_model')

# Load
loaded_model = keras.models.load_model('saved_model/my_model')

# TensorFlow Lite
converter = tf.lite.TFLiteConverter.from_saved_model('saved_model/my_model')
tflite_model = converter.convert()

Best Practices

Do:

• Use @tf.function for performance
• Use tf.data for data pipelines
• Enable mixed precision training
• Profile with TensorBoard

Don't:

• Use Python loops in tf.function
• Create tensors inside training loops
• Ignore eager vs graph mode
• Skip model validation

Troubleshooting

Issue	Cause	Solution
GPU OOM	Memory limit	Reduce batch size
Slow training	Not using GPU	Check device placement
Graph error	Incompatible shapes	Check tensor dimensions

References

• TensorFlow Documentation
• Keras Documentation