TensorFlow in 60 Minutes

TL;DR: Learn how to build machine learning models using TensorFlow in 60 minutes with hands-on examples including neural networks and structural time series forecasting.

Tutorial in 30 Seconds#

TensorFlow is an open-source machine learning framework from Google for building and training neural networks and probabilistic models.

Key capabilities:

Tensors and automatic differentiation: Immutable multi-dimensional arrays optimized for CPUs, GPUs, and TPUs with efficient gradient computation
Keras API: High-level interface for rapidly building and training neural networks
TensorFlow Probability: Probabilistic programming for Bayesian inference and uncertainty quantification
Interpretable models: Structural decomposition reveals which components drive predictions

This tutorial's goal is to show you in 60 minutes:

The core APIs of TensorFlow (tensors, variables, automatic differentiation)
How to build and train neural networks with Keras
Probabilistic modeling with TensorFlow Probability distributions and Bayesian inference

Official References#

Tutorial Content#

This tutorial includes all the code, notebooks, and Docker containers in tutorials/tensorflow

README.md: Instructions and setup for the tutorial environment
A Docker system to build and run the environment using our standardized approach
tensorflow.API.ipynb: Tutorial notebook focusing on core APIs and fundamentals
tensorflow.example.ipynb: Advanced end-to-end structural time series forecasting example
- Data Generation: We generate a realistic synthetic daily time series that combines:
- Model Building: We approximate the posterior over model parameters using Variational Inference (VI)
- Forecasting and Evaluation
tensorflow_utils.py: Utility functions