-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathreadme_tutorial
31 lines (24 loc) · 1.53 KB
/
readme_tutorial
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
------------------------------------------------------------------------------------
HELLO, WELCOME TO THE INTEL-OPTIMIZED CONTEXT IN WHICH TO RUN TENSORFLOW-PYTHON
SCRIPTS.
------------------------------------------------------------------------------------
SHORT TUTORIAL:
1) PASTE IN THE CURRENT FOLDER THE SCRIPT YOU WANT TO RUN, ALONG WITH ALL THE
OTHER REQUIRED SCRIPTS & FILES: YOU CANNOT REFERENCE STUFF OUTSIDE THIS
FOLDER.
2) NAME YOUR STARTING SCRIPT "main.py"
3) RUN "bash run.sh" ON THE COMMAND LINE AND SELECT THE REQUIRED LEVELS OF CPU
& MEMORY USAGE.
---THATS IT*--
.....................................................................................
Your Script will now be running inside a CPU-&-Memory constrained docker container
in an Intel-CPU-Optimized fashion
.....................................................................................
*TWO LAST THINGS
1. If you are using a non-standard library besides tensorflow, scikit, numpy, pandas,
keras, mtplotlib, seaborn, pickle or json then you may need to add it to the
self-explanatory file called "peat/requirements.txt". This solves the "missing library"
issue
2. Sometimes running "Docker" requires "sudo". Even tho runnng "sudo bash run.sh" might
be scary, I encourage you to open all the files and check for yourself that they are safe
----------------------------------END-------------------------------------------------