Benchmarks: A harness for running benchmarks
|
12-04-2017, 12:09 AM
Post: #1
|
|||
|
|||
Benchmarks: A harness for running benchmarks
A simple harness for selecting and running benchmarks.
You can tailor whether or not you want results printed, or in lists. (For a definitive benchmark, you wouldn't choose both). Stephen Lewkowicz (G1CMZ) https://my.numworks.com/python/steveg1cmz |
|||
12-04-2017, 12:11 AM
(This post was last modified: 12-05-2017 09:15 PM by StephenG1CMZ.)
Post: #2
|
|||
|
|||
RE: Benchmarks: A harness for running benchmarks
Version 0.1
Code:
Note: There is a complication in getting Whetstones and Benchmarks to compile. Whetstones requires OUTREAL, so Benchmarks must be compiled first... Except Benchmarks requires MWIPS so Whetstones must be compiled first. The lazy solution: 1 Comment out MWIPS 2 Compile Benchmarks 3 Compile Whetstones 4 Uncomment MWIPS and re-compile Benchmarks. *** I forgot Step 4 in Version 0.1 *** An alternative method:(with MWIPS uncommented as intended) 1 write an empty OUTREAL 2 Compile Whetstones 3 Compile Benchmarks 4 Recompile whichever OUTREAL you wish to use. Stephen Lewkowicz (G1CMZ) https://my.numworks.com/python/steveg1cmz |
|||
« Next Oldest | Next Newest »
|
User(s) browsing this thread: 2 Guest(s)