TensorFlow Tutorial

This 60-minute hands-on tutorial introduces TensorFlow and TensorFlow Probability through practical examples covering tensors, Keras 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

After this tutorial, you will understand:

• 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¶

• TensorFlow: An Open Source Machine Learning Framework
• TensorFlow GitHub repo
• TensorFlow Probability
• TensorFlow Probability GitHub repo

Getting Started¶

Prerequisites¶

This tutorial runs in a Docker container with all dependencies pre-configured. No additional setup is required beyond the steps below.

Setup Instructions¶

1. Navigate to the tutorial directory:

   > cd tutorials/TensorFlow

2. Build the Docker image:

   > ./docker_build.sh

   (See docker_build.sh)

3. Launch Jupyter Lab:

   > ./docker_jupyter.sh

   (See docker_jupyter.sh)

Dependency Management¶

This project uses uv for efficient Python dependency management within the Docker container. The system works as follows:

• requirements.in — Lists top-level package dependencies
• requirements.txt — Auto-generated pinned versions for reproducibility

The Docker container comes with all dependencies pre-compiled and synced. If you need to update dependencies manually:

# Compile top-level packages into pinned requirements
> uv pip compile requirements.in -o requirements.txt

# Sync the environment with the compiled list
> uv pip sync requirements.txt

• For more informations on the Docker build system refer to Project template readme

Tutorial Notebooks¶

Work through the following notebooks in order:

• tensorflow.API.ipynb: Core TensorFlow fundamentals

  • Tensors and tensor operations
  • Automatic differentiation
  • Keras regression models
  • TensorFlow Probability distributions

• tensorflow.example.ipynb: Advanced structural time series forecasting

  • Data Generation: Synthetic daily time series combining multiple components
  • Model Building: Approximate posterior over model parameters using Variational Inference (VI)
  • Forecasting and Evaluation: End-to-end pipeline for predictions and model assessment
  • Building trend and seasonality components
  • Incorporating holiday effects
  • Autoregressive modeling

• tensorflow_utils.py: Utility functions supporting the tutorial notebooks

Changelog¶

• 2026-03-01: Initial release