Hopefully the easiest way to get started with Google's TensorFlow Deep Learning Neural Network Python Library
Feb 7, 2016 Updated after the video
Added bash file setup-new02.sh which automates setting up ipython notebook (now called jupyter).
Installation Video at https://youtu.be/kMtrOIPLpR0
and my TensorFlow Teacher webpagge is at
install onto cloud9 http://c9.io as a custom workspace
Using the url for this repository
In the terminal type:
OR JUST RIGHT CLICK AND SELECT RUN ON THE FILE setup.sh
Note: Any of these examples can be run by just right clicking on the files and selecting run.
All files with the file name starting with rocksetta and ending with .sh should be able to be run this way.
If you have never used cloud9 it may look hard but compared to making your own linux server, cloud 9 is a breeze:
- register for a free account
- click on the big plus sign to make a new workspace
Fill out the forms as needed, the main fields are the URL for this repository (make sure it has .git at the end https://github.com/hpssjellis/easy-tensorflow-on-cloud9.git)
Make sure the default custom box is selected
- Then just right click and run setup.sh and take a break for about 10 minutes
Then hopefully try some examples in the rocksetta-example folder
My TensorFlow API diagram is at http://rocksetta.com/tensorflow-teacher/tensorflow-svg.html which is a clickable version of
My Tensorflow-teacher site is at http://rocksetta.com/tensorflow-teacher/
Other peoples examples are in the other folders.
A good starting point is the try-tf folder explained at this website
By Jeremy Ellis Maker of rocksetta.com
sets up the environment
note just type
to get your cursor back
I include this code in each of my .sh files but you could run the code in the command line and call the files normally.
to find tensorflow