What is TensorFlow?

In this post, we are going to get a basic understanding of:

• What is TensorFlow?
• What is offered in the context of machine learning?
• Tensorflow used in applications
• Types of TensorFlow

TensorFlow:

It is a free and open-source platform for high-performance numerical computation, specifically for machine learning (ML) and Deep learning (DL). It has a flexible architecture and can be deployed across a variety of platforms like CPUs, GPUs, and Google TPUs as well as mobile and edge devices.

Tensorflow in Machine Learning:

Tensorflow makes it easy to build and deploy Machine Learning solutions. So, this means that it’s not a simple framework just to build a machine learning project but a complete ecosystem that provides tools at each stage of the entire machine learning workflow:

• Data preprocessing to feature engineering
• Model training to model serving
• Data Pipelines to model inference

TensorFlow Used in Application:

• Search Engines
• Text Translation
• Image Captioning
• Interpreting
• Recommendation Systems
• Weather forecasting

A tensor is simply a typed multi-dimensional array. It can be 0 to N-dimensional.

Types of Tensorflow:

There we have different types of Tensors

Zero-dimensional: Which we call the Scalar. As you can see on the screen it has only the magnitude.

One-dimensional: which we usually call the Vector, it has the magnitude and the direction as you can see on the screen.

Two-dimensional: which we usually call a Matrix, this can be represented as a table of numbers.

Three-dimensional: which is also a matrix but it can be represented as a cube.

N-dimensional: represented as a matrix.

Great, at this stage you should be familiar with TensorFlow, what it is, why we should use it along the most basic concept of Tensors.