TensorFlow Python Library: APIs

TensorFlow offers a variety of APIs that cater to different needs, from beginners to experts in machine learning. These APIs allow for the creation, training, and deployment of sophisticated machine learning models. Here’s an overview of the main TensorFlow APIs:

TensorFlow Core API

The Core API gives you complete control over the creation and training of machine learning models. It’s particularly useful for users who want to develop custom behaviors. The Core API operates at a low level and requires detailed coding, making it suitable for advanced users who need fine-grained control over their models.

Keras API

TensorFlow’s Keras API is a high-level API that makes it easy to construct, train, evaluate, and execute all sorts of neural networks. Keras is user-friendly, modular, and extensible, which has made it very popular for both beginners and experienced users. It’s the preferred API for many common machine learning tasks and is fully integrated into TensorFlow, offering a simplified interface with functionality backed by the TensorFlow engine.


TF-Learn (or TensorFlow Learn) is a high-level Python module for building and training neural networks in TensorFlow, integrated within the TensorFlow library. It provides a higher-level abstraction for TensorFlow, making it easier to create standard models with fewer lines of custom code.

Estimator API

The Estimator API is designed to facilitate the training and evaluation of models. It allows users to create a model using a high-level abstraction, making the process of conducting training, evaluation, prediction, and export for serving simpler. Estimators are very effective for scalable and distributed training, and for models that are straightforward to deploy and manage.


TensorFlow.js is a library for developing and training ML models in JavaScript, and deploying in the browser or on Node.js. It provides WebGL-accelerated inference and training of machine learning models directly in the browser.

TensorFlow Lite

TensorFlow Lite is TensorFlow’s lightweight solution for mobile and embedded devices. It enables on-device machine learning inference with low latency and a small binary size, which is perfect for mobile phones, embedded Linux devices, and other edge devices.

TensorFlow Extended (TFX)

TensorFlow Extended is an end-to-end platform that enables the deployment of production-ready machine learning pipelines. It covers data ingestion, preprocessing, model training, serving inference, and managing deployments to online, native mobile, and JavaScript targets.

TensorFlow Hub

TensorFlow Hub is a library for the publication, discovery, and consumption of reusable parts of machine learning models. With TensorFlow Hub, you can share or reuse pretrained components of a TensorFlow model along with their weights and implementation.