Python Neural Networks - Tensorflow 2.0 Tutorial - Loading & Looking at Data

This python tensorflow 2.0 tutorial covers how to load in the MNIST Fashion dataset that our neural network will use for image classification in future videos. We can load the data set quite easily by using the keras API provided by tensorflow. We will then use matplotlib to view our data.

Tensorflow install: pip install -q tensorflow==2.0.0-alpha0

Want a sneak peak into my life? Follow my Instagram @tech_with_tim where I'm going to be filming a video each morning sharing my goals for the day and what I have planned:

◾◾◾◾◾
💻 Enroll in The Fundamentals of Programming w/ Python

◾◾◾◾◾◾

⚡ Please leave a LIKE and SUBSCRIBE for more content! ⚡

Tags:
- Tech With Tim
- Tensorflow Tutorial
- Python tensorflow 2.0 tutorial
- Python neural network turtorial
- Python Tutorials
- Keras MNIST

[konventionmedia]

I quit varsity to learn on my own and Thank God I found Tim... all the way from South Africa

[SzalonyOnufry]

Tim thank you so much for this tutorial, I wanted to create deep learning project for a long time but I was putting it off because I didn't know where to start after only doing tutorials that required me to fill in some code, not to write it from scratch. You are showing the process step by step and focusing on things that no other tutorials focuses on, like how to look at data which help me a lot as I'm new to python. Best tutorial I've ever seen <3

[PC_Pete]

Can already tell I'm going to learn a lot from this series. Can't wait for more!

[abdullah]

As always tim. master of teaching someone anything. I have taken your pygame tutorials and watched your stream of 12 hours. Man o man it was so exciting and awsome but i made a mistake i watched the stream before i took your tutoirals of pygame😂😂
But even if i didn't have a clue about what your doing but i watched it and my heart was saying its going to be something amazing..
Great work over all..
Good Job

[SinghFlex]

Your tutorials and explanations are straight to the point. Respect from INDIA brother !

[JanRutger]

Hi Tim, thank you for all the great videos made especially for us.
This afternoon I installed Tensorflow 2.x on my Raspberry and used the program from this video to test Tensorflow on my Pi.

It worked all at once, a good starting point to continue now. and thanks again

[TheoBeats]

This is such a good series of videos so far! You're really good at teaching Tim!

[vivekpatil]

excellent video. just a small correction at 11:08 -> 255 is all ones in binary and hence it is known as white and 0 is black.

[Herebuss]

Thanks brew. Nice pace and explanation, allowing us to play along.

[anilkumar-cuwe]

i like ur way off teaching for sure i will honor you

[erfanmonemi]

Thank so much for your awesome course and other videos. Your work is great, keep on.

[deivcohen]

A small correction, in imshow the image should be displayed in grayscale ( cmap='gray' ) and not in binary which simply turns the image into 0, 1.

[madigor]

Cant wait till you explain deep q learning

[MHatter]

I was having an issue where the images where not being displayed when calling plt.show(). In the hope of saving someone else some time, matplotlib does not like virtualenv. If, like me, you want do this in a virtual environment, make sure you use Python’s built in venv module.

[fadhelalsuwaidi]

Hi guys
I tried to install tensorflow but it doesn't works even after I uninstall numpy, this message is shown:
ERROR: Could not find a version that satisfies the requirement tensorflow==2.0.0-alpha0 (from versions: none)
ERROR: No matching distribution found for tensorflow==2.0.0-alpha0


Help please.

[vireshuberoy]

Could you also please make a video explaining how to write good and clean code?

[georgemarinop]

Tim, awesome work so far .Thank you so much for sharing all that knowlegde ...
Now, in this tutorial i'm facing a problem while running the file at 08:00 . The exact message that pops up is ...
Traceback (most recent call last):
File "NNTutorial#1.py", line 1, in <module>
import tensorflow as tf
File "C:\Users\georg\anaconda3\lib\site-packages\tensorflow\__init__.py", line 27, in <module>
from tensorflow._api.v2 import audio
ModuleNotFoundError: No module named 'tensorflow._api' .
I would be so grateful if anyone can help me .

[SMFahim-vozn]

If train_images[0].size equals (28, 28, 3), then 3 represents RGB. If 4, then RGBA

[brianm.]

Hi Tim. Noob here. Question about your methodology in scaling down the pixel data values by 255. Why did you opt for floats instead of integers? I thought integers required less resources than floats to process. Thanks!

[huynhatnguyen]

hello, i'm having trouble importing the fashion_mnist dataset, but it gets weird when i try to load another dataset like boston_housing, it still declares that it can not import 'fashion_mnist' name from 'datasets'.