Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Categories
- All Discussions1,524
- General534
- Graphics109
- GPU Computing419
- Mobile141
- Pro Graphics163
- Tools158
Tags in this Discussion
- cuda 422
- gpu-debugging 81
Could someone please help me compile this CUDA code that works?
-
Hey everyone, I am fairly new to GPU programming, and have spent many fruitless hours trying to compile some dimension reduction code that uses CUDA and Matlab. I was wondering if anyone here who already has a system configured with Matlab and CUDA would be willing to quickly build the source for me (which is a 255k zip file) and send me a compiled executable. I have corresponded with the author of the algorithm, and while he couldn't help me with the compiling, he also expressed an interest in having the compiled version to offer to others on his website-- so you would be doing a service to the community of researches in the machine learning community. Here is a link to the file:
where it says "CUDA implementation":
http://homepage.tudelft.nl/19j49/t-SNE.html
In case the above link isn't working:
http://db.tt/B4Wq6P6P
Thanks-- any help will be greatly appreciated! -
5 Comments sorted by
-
Why did the author of the algorithm say it won't compile?
Have you considered using Jacket for this? Jacket has very fast reductions (see benchmarks under "Utility"). -
The author uses linux personally and is not able to compile a windows version. Yes, I have a license for Jacket already, but this is hand tuned CUDA code that is called through a binary MEX file, and that's what I am using because it already exists and is fast. Thanks.
-
I reached out to the author, Laurens, and he has been gracious enough to let us include the code as an example in Jacket. We'll get that queued up soon and then you'll have it natively in Jacket. Cheers!
-
Wow, that's unbelievable. Thanks so much! Can't wait to see what the code looks like. Will this be included in one of the nightly builds first? Thanks again.
-
I'd actually still like to try to compiled native CUDA code-- it will be interesting to benchmark it against Jacket with Matlab. If there isn't much of a speed difference, then it is a pretty good argument that, for prototyping purposes--assuming you have a license for Jacket-- you're better off just doing things that way versus writing your own CUDA code in C. Anyway, perhaps one of the engineers from AccelerEyes could take a quick look at compiling the author's code. Thanks again.
