TensorFlow by Google: What is it and how do I know?

Google TensorFlow: H ή Machine learning είναι μια τεχνολογία υπολογισμού που εξελίσσεται συνεχώς τα τελευταία χρόνια. Η συγκεκριμένη τεχνολογία χρησιμοποιείται παντού γύρω μας, από την οδήγηση αυτοκινήτων μέχρι την πρόβλεψη της χρηματιστηριακής αγοράς.TensorFlow

TensorFlow is a Google program based on machine learning and neural networks. Below we will look at exactly what can be used and how to learn to use it.

What is TensorFlow?

To explain what TensorFlow is, we must first understand what mechanical learning is. Mechanical learning and neural networks already affect our lives in much more ways than you think.

Machine learning could be said to be the process of teaching computers to learn how to analyze data and make informed decisions without being directly programmed to do so.
In order to do this, we train neural networks that can perform specific tasks.

TensorFlow is an open source neural network library from Google, developed by the Google Brain team for many uses. Essentially, TensorFlow removes the need ς ενός νευρικού δικτύου από την αρχή. Έτσι αφού υπάρχει ήδη η , you can train TensorFlow with your own data and use the results you want.

Does it seem abstract to you? What can you do with a neural network? Let's look at some examples of TensorFlow

Sorting pictures

Many training seminars for novice users use image sorting to help understand. By providing reference pictures on a neural network, the network can learn (mechanically) whether an image contains similar objects.

To see this process in action, take a look at Darth Vader classifier by Siraj Raval.

This kind of assisted sorting of data from the computer is incredibly powerful and not just the way it is used in the video. TensorFlow is already used in biomedical image analysis.

Almost every field based on the analysis of large amounts of image data can benefit from this technology. As shown in the official TensorFlow presentation video, it's used in medicine, art, Gmail, or wherever you can imagine.

Deep Photo Style Transfer

In addition to image sorting, TensorFlow can be used to dynamically change images. Deep Photo Style was developed by a team at University of Cornell. The project takes two different photos and creates a new style image from both original, with amazing results.

Magenta AI Music

What can you say about creating artworks? With nerve network libraries like TensorFlow, it can become reality.
Magenta uses TensorFlow to create tools for musicians. Using deep learning, Magenta creates the right tools that musicians need for new sound mixing, and much more.

By creating improvised musical pieces created by the nervous network, the Magenta allows anyone to create unique and beautiful sounds without having any musical knowledge.

How to Learn TensorFlow

Machine learning is not easy. We need a good understanding of statistics, math, programming and general data science, as all of the above are necessary to teach a machine to "think" and make decisions. But TensorFlow offers lessons even for beginners. The official tutorials TensorFlow guides you step by step for every setting and every use.

Most TensorFlow projects use the Python programming language. If you do not know it, there are many websites that help you learn Python (Dive Into Python, Official Python Tutorial, TryPython ή LearnPython). If you already know JavaScript, TensorFlow has video tutorials for the TensorFlow.js library.

These lessons, along with Google's free mechanical learning, will help you understand every project.

TensorFlow is an incredibly powerful tool from the biggest company on the web. The decision to make it available as open , makes technology accessible to everyone (those willing to learn).

__________________________________

iGuRu.gr The Best Technology Site in Greecefgns

every publication, directly to your inbox

Join the 2.100 registrants.

Written by giorgos

George still wonders what he's doing here ...

Leave a reply

Your email address is not published. Required fields are mentioned with *

Your message will not be published if:
1. Contains insulting, defamatory, racist, offensive or inappropriate comments.
2. Causes harm to minors.
3. It interferes with the privacy and individual and social rights of other users.
4. Advertises products or services or websites.
5. Contains personal information (address, phone, etc.).