Tensor-flow Quick Overview

FLOW

So, what is Tensor-Flow about?

Tensor Flow is a fast, flexible, and scalable open-source machine learning library for research and production

Okay that sounds and looks complicated, for better understanding lets divide and conquer

Tensor-In a simplistic term a Tensor can be compared to data storage drive

When we have yottabyte of data (As of now yottabyte space of data storage isn’t feasibly achieved) and we want to store the data across multiple data storage capacity devices

In case of 2-dimeneionsial, 1 dimensional and point tensors let’s consider we have 64TB ,128 TB and 256 TB storage devices, then we will be assigning 64 TB storage device rank-0, 128TB storage device rank-1 and 256TB storage device rank-2

What’s with all these dimensions and Ranks?

Dimensions indicate the data holding capacity of Tensors just as with Ranks we assigned 64 TB rank-0,128TB rank-1and 256 TB rank-1 based on storage (in case of Tensors its dimensions) capacity

FLOW- Blue print of steps to follow to accomplish a Task, Okay I Guess this is over-simplifying

Okay let’s think for a moment that you are to do research for your boring and monotonous assignment of 10 marks in 30 minutes and the marks are not dividend across uniformly among all the questions

Well for any task there will be work-flow and in Tensor Flow context the data stored in tensors are computed with a specific flow to achieve the end goal.

Why Tensor Flow?

Let’s Think for a minute why should you spend time learning and understanding about Tensor Flow Why should you understand about Tensor Flow? why should consider implanting your project with Tensor Flow?

Tensor Flow offers multiple levels of abstraction so you can choose the right one for your needs. Build and train models by using the high-level Kera’s API, which makes getting started with Tensor Flow and machine learning easy.

Tensor Flow Stream-lines the Complete cycle of Building Machine Learning Models from Data extraction to deployment

When Should you consider Bringing Tensor Flow on Board?

Text based applications such as sentimental analysis (CRM, Social Media), Threat Detection (Social Media, Government) and Fraud Detection (Insurance, Finance)

Text Summarization

Summarization can be learned with a technique called sequence-to-sequence learning. This can be used to produce headlines for news articles. And briefing review of a product.

Image Recognition

Mostly used by Social Media, Telecom and Handset Manufacturers; Face Recognition, Image Search, Motion Detection, Machine Vision and Photo Clustering can be used also in Automotive, Aviation and Healthcare Industries.

Written by

Siddhartha kancharla