# Deep Learning with Python, TensorFlow, and Keras tutorial

///Deep Learning with Python, TensorFlow, and Keras tutorial

## Deep Learning with Python, TensorFlow, and Keras tutorial

An updated deep learning introduction using Python, TensorFlow, and Keras.

Text-tutorial and notes:

TensorFlow Docs:
Keras Docs:

Discord:

source

By |2019-06-18T19:53:46+00:00June 18th, 2019|Python Video Tutorials|44 Comments

1. robin jaison June 18, 2019 at 7:53 pm - Reply

Nice video!!

2. armer18 June 18, 2019 at 7:53 pm - Reply

Is it possible to use tensorflow on integrated graphics card (i5-8250U)?

3. Alex Joslin June 18, 2019 at 7:53 pm - Reply

Thought you where Edward Snowden at first.

4. harisimer June 18, 2019 at 7:53 pm - Reply

Can you explain what y_train and x_train exactly is?

5. Aakarshan Chauhan June 18, 2019 at 7:53 pm - Reply

what about x_train = x_train/255.0 ? that will also give us a value between 0 and 1 with max being 1.0
so can we use that instead of normalise? or are they the same?

6. Boaz Nathan June 18, 2019 at 7:53 pm - Reply

Great video!

7. Nguai al June 18, 2019 at 7:53 pm - Reply

is this min max normalization?

8. MarkOates June 18, 2019 at 7:53 pm - Reply

Thank you! Great video and walkthrough.

9. Julien M June 18, 2019 at 7:53 pm - Reply

Why 128 ? This is 128 neurones? The input is an image of 28×28? So why?

10. Zhen Xiang June 18, 2019 at 7:53 pm - Reply

Great tutorial! Covers almost everything and easy to understand. But you may want to reconsider how you normalize the pixel values. The pixel values before normalization are in the range of [0, 255] for all the images, and we hope the maximum pixel value in EACH image to be 1 after normalization. Otherwise, we may get troubles if a compact network structure is used for training. So you may want to consider x_train /= 255.0 instead. (Please ignore my comment if I was wrong:))

11. sosscs June 18, 2019 at 7:53 pm - Reply

can you do a tutorial on word2vec with neural networks

12. Shaina June 18, 2019 at 7:53 pm - Reply

Great video thank you so much!

13. Evdokimos Theodoridis June 18, 2019 at 7:53 pm - Reply

Your tutorials are great (big fan here) ,but i faced a problem with the model.evaluate, when run it, displays this error "ValueError: Input arrays should have the same number of samples as target arrays. Found 60000 input samples and 10000 target samples."

14. Mai Mariarti June 18, 2019 at 7:53 pm - Reply

You lost me after 5:50
Do you have a more basic playlist that shows how to get into AI and ML?
I have no idea what TensorFlow and Keras do.

15. TheArchit1 June 18, 2019 at 7:53 pm - Reply

Bro I Love You (NO HOMO)

I gotta get that shark mug
you just awesome with your shark cup in the same screen！

17. mohamed ali June 18, 2019 at 7:53 pm - Reply

Well done, thank you!
Finally what I have been waiting for..

18. k shamanth kumar June 18, 2019 at 7:53 pm - Reply

This is for a single digit recognition right what about the reading multi digit recognition at the same time ??

19. esmail59 June 18, 2019 at 7:53 pm - Reply

Well done, thank you!

20. Antony Mapfumo June 18, 2019 at 7:53 pm - Reply

Awesome tutorial. Much appreciated.

21. Phil S June 18, 2019 at 7:53 pm - Reply

Hi everyone. Just not clear on how to get owncloud or nextcloud etc, to have projects run as a function to train models with out having to enter commands. I like have a datacenter like system and able to run jobs. Be safe Phil

Sir kindly make videos on these contents which are under mentioned???
Building AI systems through Deep Learning and NVIDIA GPUs: Keras and

Tensorflow

Module Contents:

• What is Deep Learning and what are Neural Networks?

• Artificial Neural Networks (ANN) Intuition

• Building an ANN

• Evaluating Performance of an ANN

• Hands-On Exercise

• Introduction to Keras and TenserFlow

• Convolutional Neural Networks (CNN) Intuition

• Building a CNN

• Evaluating Performance of a CNN

• Hands-On Exercise

• Recurrent Neural Networks (RNN) Intuition

• Building a RNN

• Evaluating Performance of a RNN

• Hands-On Exercise

• Image Classification with DIGITS

• Object Detection with DIGITS

• Neutral Network Deployment with DIGITS and TensorRT

23. PRITAM SARKAR June 18, 2019 at 7:53 pm - Reply

why 128 neurones ????

24. Denis Denisenko June 18, 2019 at 7:53 pm - Reply

terrible teacher

25. Bot June 18, 2019 at 7:53 pm - Reply

why is there no comments on that ridiculous yet awesome cup

26. Lantic Kin June 18, 2019 at 7:53 pm - Reply

you just awesome with your shark cup in the same screen！

27. Mickelodian Surname June 18, 2019 at 7:53 pm - Reply

Is that digital oceans Sammy the shark I seen you drinking from?

28. コペラマルツェル June 18, 2019 at 7:53 pm - Reply

29. Anish Kelkar June 18, 2019 at 7:53 pm - Reply

This tutorial gave a hands on approach to machine learning unlike most of the other tutorials. Thank you very much.

30. BUBAI NAYAK June 18, 2019 at 7:53 pm - Reply

Hi @sentdex could you please give me some idea to trained image model for image processing using kerars to run in raspberry pi

31. sagar lama June 18, 2019 at 7:53 pm - Reply

I have no idea what hes doing but I'm still following hoping that some day after rewatching this video, it'll make sense

32. Arthur Tabbal June 18, 2019 at 7:53 pm - Reply

Thanks for this video ! It changed my life ! A question.. is it always better to do deep learning with tensorflow over that sci kit learn "simple" machine learning iris set thing ? Whats the real difference?

33. Waseem Ali June 18, 2019 at 7:53 pm - Reply

How can I decide to pass 128 units like you did in adding 128. You said this is number of neurons. How can I decide the number of neurons according to the data I have.

34. Abrar Akib June 18, 2019 at 7:53 pm - Reply

Which GPU are you using in this tutorial?

35. sc13nt1st June 18, 2019 at 7:53 pm - Reply

val_loss, val_acc = model.evaluate(x_test, y_test)

print(val_loss, val_acc)

Results in an error

36. Manar Mansour June 18, 2019 at 7:53 pm - Reply

I just have a question .. why do we normalize the data using axis=1 instead of any other axis ??

37. chinggis 666 June 18, 2019 at 7:53 pm - Reply

man how install tensorflow library ?

38. Spacetime Matter June 18, 2019 at 7:53 pm - Reply

How can i see as a picture which feature was extracted by this neural network.I always see picture in the papers is changing in every layer.And they tell firts layer find this features second layer find these features.But i can i see my networks find.

39. KOlda June 18, 2019 at 7:53 pm - Reply

Hey, why the model is not working if I use any other loss function? I really don´t see the reason. Thanks for answer.

40. Eic Diamon June 18, 2019 at 7:53 pm - Reply

41. Alina Aldea June 18, 2019 at 7:53 pm - Reply

Hi! I have a dataset (2 columns) containing a text and the asscoiated score. i'm trying to do a sentiment analysis using tensorflow, but any suggestion on how I can start? I'm reading my csv in python but then i get stuck, i'm so new to this. Any help is much appreciated!

42. Gru June 18, 2019 at 7:53 pm - Reply

I gotta get that shark mug

43. Emil Eremiev June 18, 2019 at 7:53 pm - Reply

Hi sentdex, the explanation of this video is excellent! btw… how you decide to use two hidden layers with 128 neurons ?

44. Frgh Nsrp June 18, 2019 at 7:53 pm - Reply

Thanks for the great tutorial, was just wondering how you decide the number of neurons? Thanks.