Get Started with TensorFlow

TensorFlow is an open-source machine learning library for research and production. TensorFlow offers APIs for beginners and experts to develop for desktop, mobile, web, and cloud. See the sections below to get started.

Learn and use ML

The high-level Keras API provides building blocks to create and train deep learning models. Start with these beginner-friendly notebook examples, then read the TensorFlow Keras guide.

  1. Basic classification
  2. Text classification
  3. Regression
  4. Overfitting and underfitting
  5. Save and load
import tensorflow as tf
mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(),
  tf.keras.layers.Dense(512, activation=tf.nn.relu),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)
Run in a Notebook

Research and experimentation

Eager execution provides an imperative, define-by-run interface for advanced operations. Write custom layers, forward passes, and training loops with auto‑differentiation. Start with these notebooks, then read the eager execution guide.

  1. Eager execution basics
  2. Automatic differentiation and gradient tape
  3. Custom training: basics
  4. Custom layers
  5. Custom training: walkthrough
  6. Example: Neural machine translation w/ attention

ML at production scale

Estimators can train large models on multiple machines in a production environment. TensorFlow provides a collection of pre-made Estimators to implement common ML algorithms. See the Estimators guide.

  1. Build a linear model with Estimators
  2. Wide and deep learning with Estimators
  3. Boosted trees
  4. How to build a simple text classifier with TF-Hub
  5. Build a Convolutional Neural Network using Estimators

Google Colab: An easy way to learn and use TensorFlow

Colaboratory is a Google research project created to help disseminate machine learning education and research. It's a Jupyter notebook environment that requires no setup to use and runs entirely in the cloud. Read the blog post.

Build your first ML app

Create and deploy TensorFlow models on web and mobile.

Web developers

TensorFlow.js is a WebGL accelerated, JavaScript library to train and deploy ML models in the browser and for Node.js.

Mobile developers

TensorFlow Lite is lightweight solution for mobile and embedded devices.

Videos and updates

Subscribe to the TensorFlow YouTube channel and blog for the latest videos and updates.

Get started with TensorFlow's High-Level APIs

Eager execution

tf.data: Fast, flexible, and easy-to-use input pipelines