Neural Networks and TensorFlow - Deep Learning Series [Part 11]

in #deep-learning8 years ago (edited)

Neural Networks and TensorFlow - 11 - Logistic Regression.png

In this lesson you're going to learn how to implement logistic regression using TensorFlow in Python. Implementing linear regression was relatively easy.

Logistic regression is not hard either, nor does it involve something much different than linear regression. In logistic regression, we're simply using a non-linearity for activation. To be more precise, we're using a non-linear function for activation.

Most often, this non-linearity is a sigmoid, a hyperbolic tangent, or a rectified linear unit. What's great about TensorFlow is that it comes with methods to simply call these functions, without actually hard coding them. So, it could either be:

  • tf.sigmoid
  • tf.tanh
  • tf.nn.relu

And this makes the implementation much more convenient and easier. Of course, you're better off if you know what these function do and how they look under the hood, but not everyone needs to know this in order to work with TensorFlow. Anyhow, please see the video below for doing logistic regression in TensorFlow.



To stay in touch with me, follow @cristi


Cristi Vlad Self-Experimenter and Author

Sort:  

Still hurts my head little :-)
Jokes aside, I am actually using your videos to do some revision. Thanks for the upload.

Hey, what kind of revision?

I started working again around last year and haven't really been in touch with ANN, machine learning and Programming for even more than that.
The last 2-3 neural n/w videos on tensor flow has served as a reminder for a lot of stuff that I had dropped, even from my own personal reading.

Really helpful and informative. Great explanation. Can yu maybe have a series on the chem programme, Aspen soon also?

whats chem?

Incredible post, too informative and very easy to understand, you've definitely been here for a long time here at Steemit doing a great job and I hope you keep growing more and more, you are one of my favorite users, many greetings @cristi!

Imi face placere sa te vad printre cei votati de SteemSTEM.
Am mai urmarit seria de articole, din pacate nu pot tot timpul sa ma uit la video. Daca ai putin mai mult timp, poti intra pe canalul de Discord al SteemSTEM si poti sa iti cresti expunerea blogului si asa. Prin socializare.
Spor!

am fost unul dintre membrii care au inceput steemstem, dar am plecat pe drumuri diferite la un moment dat.

De asta te voteaza lemouth pe unde te gaseste :p
Foarte tare, am incercat sa mai aduc membri romani, dar sunt reticenti. Ma refer la cei care ar putea scrie calitativ. Pe mine m-au promovat de curand si ii mai ajut cu engagementul si mai sunt mentor.
Trebuie sa impac si steemromania si steemstem si munca :)
Respect, te mai urmaresc! Programator sunt la baza..

Oricine scrie posturi de calitate (stiintifice si originale) va fi votat. In ce programezi? :)

C#, C++
Fac si modelare 3D si putin engine grafic si shadere.

Congratulations @cristi! You have completed some achievement on Steemit and have been rewarded with new badge(s) :

Award for the number of comments received

Click on any badge to view your own Board of Honor on SteemitBoard.

To support your work, I also upvoted your post!
For more information about SteemitBoard, click here

If you no longer want to receive notifications, reply to this comment with the word STOP

Upvote this notification to help all Steemit users. Learn why here!