TensorFlow™ is an open source software library for high-performance numerical computation. Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices. Originally developed by researchers and engineers from the Google Brain team within Google’s AI organization, it comes with strong support for machine learning and deep learning and the flexible numerical computation core is used across many other scientific domains.
It is compatible with frameworks like CUDA pandas, scikit-learn, Theano, Keras, numpy, scipy etc.
Tensorflow has been promoted and supported well compared to other libraries and is scalable for on-demand computational requirements. Cloud service providers like Google cloud, Amazon and Azure provide tensorflow as service, that can help organizations to speed up development and deployment, in a secure environment.
How can one use Tensorflow in Machine Learning?
Tensorflow in itself is a library for high-performance numerical computation. One has to use Keras or Theano or Scikit in combination with Tensorflow to build Machine Learning applications. Mostly application is written in python, but other programming languages can also be used using different wrapper libraries.
The high-level Keras API provides building blocks to create and train deep learning models.
How to Install Tensorflow
Tensorflow comes in two flavors, one that use CPU and one that uses CUDA and GPU. Depending on the hardware, you can choose the flavor you want to use.
Python should be installed in the system, preferably with 64-bit processor.
A detailed installation guide is available at https://www.tensorflow.org/install/
Tensorflow is written in C and the C API is wrapped using different libraries to support different programming languages.
A detailed reference to architecture is available at https://www.tensorflow.org/guide/extend/architecture
The complete source code of tensorflow is available at https://github.com/tensorflow/tensorflow
We recommend using the package provided from https://www.tensorflow.org/install/ for use of application development.