TFX tutorials | TensorFlow

TensorFlow in Production tutorials

Caution: TFX is currently experiencing issues with Python 3.10 in Colab.

To run the tutorials, modify the notebook as follows:

Add a new code cell at the top of the notebook.
Enter this code in the cell:
%%shell sudo apt update sudo apt install python3.8 python3.8-distutils python3.8-dev update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.8 3 curl -O https://bootstrap.pypa.io/get-pip.py python get-pip.py
Replace the code in the cell which installs TFX with this code:
# 1) TFX relies on an old version of google-api-core so we let google-auth float # for the install. We grep it out below: !grep -v google-auth /etc/requirements.core.in > requirements.txt # 2) httplib2 should be included in /etc/requirements.core.in but it's not for # reasons. We ensure it's included: !grep httplib2 /etc/requirements.user.in >> requirements.txt # 3) google.colab package is not available as a wheel. We symlink that in so # it's on the sys.path of Python 3.8: !mkdir /usr/local/lib/python3.8/dist-packages/google !ln -s /usr/local/lib/python3.10/dist-packages/google/colab /usr/local/lib/python3.8/dist-packages/google/colab # Now with those pre-requisites out of the way: !pip install tfx==1.13.0 -r requirements.txt

TFX 1.0

We are happy to announce the availability of the TFX 1.0.0. This is the initial post-beta release of TFX, which provides stable public APIs and artifacts. You can be assured that your future TFX pipelines will keep working after an upgrade within the compatibility scope defined in this RFC.

Getting started tutorials

1. Starter Pipeline

Probably the simplest pipeline you can build, to help you get started. Click the Run in Google Colab button.

2. Adding Data Validation

Building on the simple pipeline to add data validation components.

3. Adding Feature Engineering

Building on the data validation pipeline to add a feature engineering component.

4. Adding Model Analysis

Building on the simple pipeline to add a model analysis component.

TFX on Google Cloud

Google Cloud provides various products like BigQuery, Vertex AI to make your ML workflow cost-effective and scalable. You will learn how to use those products in your TFX pipeline.

Running on Vertex Pipelines

Running pipelines on a managed pipeline service, Vertex Pipelines.

Read data from BigQuery

Using BigQuery as a data source of ML pipelines.

Vertex AI Training and Serving

Using cloud resources for ML training and serving with Vertex AI.

TFX on Cloud AI Platform Pipelines

An introduction to using TFX and Cloud AI Platform Pipelines.

Next steps

Once you have a basic understanding of TFX, check these additional tutorials and guides. And don't forget to read the TFX User Guide.

Videos and updates

Subscribe to the TFX YouTube Playlist and blog for the latest videos and updates.

TensorFlow in Production tutorials

TFX 1.0

Getting started tutorials

1. Starter Pipeline

2. Adding Data Validation

3. Adding Feature Engineering

4. Adding Model Analysis

TFX on Google Cloud

Running on Vertex Pipelines

Read data from BigQuery

Vertex AI Training and Serving

TFX on Cloud AI Platform Pipelines

Next steps

Complete Pipeline Tutorial

Custom Component Tutorial

Data Validation

Model Analysis

Serve a Model

Videos and updates

TFX: Production ML with TensorFlow in 2020

TFX: Production ML pipelines with TensorFlow

Taking Machine Learning from Research to Production