Getting Started
For a brief overview of TensorFlow programming fundamentals, see the following guide:
MNIST has become the canonical dataset for trying out a new machine learning toolkit. We offer three guides that each demonstrate a different approach to training an MNIST model on TensorFlow:
- MNIST for ML Beginners, which introduces MNIST through the high-level API.
- Deep MNIST for Experts, which is more-in depth than "MNIST for ML Beginners," and assumes some familiarity with machine learning concepts.
- TensorFlow Mechanics 101, which introduces MNIST through the low-level API.
For developers new to TensorFlow, the high-level API is a good place to start. To learn about the high-level API, read the following guides:
- tf.estimator Quickstart, which introduces this API.
- Building Input Functions with tf.contrib.learn, which takes you into a somewhat more sophisticated use of this API.
- Logging and Monitoring Basics with tf.contrib.learn, which explains how to audit the progress of model training.
- Exporting a Trained Model for Serving, which shows how to save a trained model in a form that is ready to deploy.
TensorBoard is a utility to visualize different aspects of machine learning. The following guides explain how to use TensorBoard:
- TensorBoard: Visualizing Learning, which gets you started.
- TensorBoard: Embedding Visualization, which demonstrates how to view and interact with high-dimensional data, such as embeddings.
- TensorBoard: Graph Visualization, which explains how to visualize the computational graph. Graph visualization is typically more useful for programmers using the low-level API.