TensorFlow is a computational framework created by Google and has become one of the most diffused deep learning toolkits. It can work with both CPUs and GPUs and already implements most of the operations and structures required to build and train a complex model. TensorFlow can be installed as a Python package on Linux, macOS, and Windows (with or without GPU support). However, I suggest you follow the instructions provided on the website (the link can be found in the info box at the end of this chapter) to avoid common mistakes and install it in the best way considering every specific environment.
The main concept behind TensorFlow is the computational graph or a set of subsequent operations that transform an input batch into the desired output. In the following diagram, there's a schematic representation of a graph:
