In this chapter, we looked at various extensions of TensorFlow for improving the productivity of data scientists and enabling the easier deployment of cutting-edge models in production at a large scale.
We looked at TensorFlow Hub, which is similar to the GitHub repository of trained deep learning models from various areas like Computer Vision, Natural Language Processing, and so on. Thereafter, we understood how TensorFlow Serving provide tools and libraries to deploy deep learning models at scale. Lastly, we learned about the open source components of TensorFlow Extended (TFX), which is a machine learning platform from Google. TFX helps in the entire model building pipeline, from data analysis to model deployment.
Next, we learned about several best practices when building scalable AI products. Building a robust engineering pipeline and...
