darknet

mirror of https://github.com/pjreddie/darknet.git synced 2023-08-10 21:13:14 +03:00

Go to file

Zeno Gantner 493b1e3c65 option to compile with -march=native to speed up CPU performance

On my laptop without GPU this gives me a significant improvement.
From about 26 seconds (9 with OpenMP) down to (still) 12 seconds (5 with OpenMP).

Intel(R) Core(TM) i5-6500T CPU @ 2.50GHz (4 cores)
gcc (Ubuntu 5.4.0-6ubuntu1~16.04.5) 5.4.0 20160609

Deactivated by default to not break e.g. distributed compilation and cross compilation.

2017-12-19 16:09:04 +01:00

cfg

it's raining really hard outside :-( :rain: :storm: ☁️

2017-10-17 11:41:34 -07:00

data

...and tree :tree:

2017-01-04 05:05:08 -08:00

examples

🌃 🤔 💭 👹

2017-11-26 12:27:21 -08:00

include

OK SHOULD I START WORKING ON CVPR OR WHAT?

2017-11-07 16:10:33 -08:00

python

I WISH I HAD SOME TESTS THOUGH

2017-11-21 11:34:46 -08:00

scripts

coco dataset script

2017-05-03 15:23:03 -07:00

src

🌃 🤔 💭 👹

2017-11-26 12:27:21 -08:00

.gitignore

🐍 🐍 🐍 🐍

2017-06-08 13:47:31 -07:00

LICENSE

Update LICENSE

2016-07-29 13:11:41 -07:00

LICENSE.fuck

YO DAWG, I HEARD YOU LIKE LICENSES

2017-06-18 13:16:16 -07:00

LICENSE.gen

generate own license, totally legal :verified: 🐍 🐍 🐍

2017-06-20 15:35:29 -07:00

LICENSE.gpl

YO DAWG, I HEARD YOU LIKE LICENSES

2017-06-18 13:16:16 -07:00

LICENSE.meta

generate own license, totally legal :verified: 🐍 🐍 🐍

2017-06-20 15:35:29 -07:00

LICENSE.mit

YO DAWG, I HEARD YOU LIKE LICENSES

2017-06-18 13:16:16 -07:00

LICENSE.v1

generate own license, totally legal :verified: 🐍 🐍 🐍

2017-06-20 15:35:29 -07:00

Makefile

option to compile with -march=native to speed up CPU performance

2017-12-19 16:09:04 +01:00

README.md

Update README.md

2015-08-05 16:56:24 -07:00

README.md

#Darknet# Darknet is an open source neural network framework written in C and CUDA. It is fast, easy to install, and supports CPU and GPU computation.

For more information see the Darknet project website.

For questions or issues please use the Google Group.

Languages

C 90.4%

Cuda 7.8%

Python 0.9%

C++ 0.4%

Makefile 0.3%

Other 0.2%