Note: This post was originally published on AH’s Blog (WordPress) on April 12, 2016, and has been migrated here.

TensorFlow is an open-source library created by Google for deep learning tasks. It represents matrix operations and data dependencies as a computation graph: edges (called tensors) carry data between operation nodes.

Key advantages of TensorFlow include reducing development time by abstracting mathematical operations and offering GPU support.

As part of a machine learning master’s course assignment, I implemented a solution for the XOR learning problem using TensorFlow — my first time using the library. The model structure is shown below:

Notebook: xor_learning.ipynb on GitHub