Hi,
I want to run my computationally expensive AI algorithms on gpu. PyOpenCL
is the best option for me coz all my AI code is in python. I have never
used GPUs before. I dont know how to learn pyOpenCL . The official
documentation assumes i know openCL i guess. Should I first learn OpenCL
and then
Hi,
PyOpenCL is awesome and its been around since 2011. Its extremely easy to
program but frustrating in the beginning. There is absolutely no
comprehensive detailed documentation explaining PyOpenCL to beginners who
have never programmed GPUs before.
Should we start drafting a simple easy to under
Hi,
Is there a detailed description of how to create a docker image for python
code that uses pyopencl and few other libraries.
Procedures for creating a docker file and running that file to create a
container are well documented, but i am struggling to find out how to write
a docker file which wou
Hi,
I installed Jupyter notebook through anaconda3 package. It comes with all
required python modules.
when i do "import pyopencl" i get an error saying no module found.
So i did
"import sys
sys.path.append('/Users/xyz/')"
where i have a pyopencl folder extracted from the tar zip file i downloaded
like conda/pip.
>
>
> - Sam
>
>
>
> On Mar 28, 2017, 4:59 AM -0400, aseem hegshetye ,
> wrote:
>
> Hi,
> I installed Jupyter notebook through anaconda3 package. It comes with all
> required python modules.
> when i do "import pyopencl" i get
Hi,
Does pyopencl have inbuilt memory cleaning. Does "pyopencl.enqueue_copy"
clear corresponding memory ( memory allocated to the object that was copied
to CPU via enqueue command) from the GPU or is the data still in the GPU
and gets cleaned only after some other process.
Is there a "pyopencl.cl
Hi,
I am going to load 3 arrays from CPU to GPU 4 times in a loop in my
function. Then GPU will work on those arrays and generate output.
1 of the 3 arrays is going to be the same every time. SO i want to find a
way to keep that 1 array *(arrazy_z)* constant instead of loading it each
time. I als
Hi,
How to release read_only & write_only memory in GPU.
suppose i load a write only array and another read only array in GPU. Then
later once I am done retrieving the output from GPU, i want to load new
arrays in the GPU. But i want to get rid of old arrays in the GPU.
This would help me when I
Hi,
Is there a way to debug and/or print intermediate variables while running a c
kernel on GPu via pyopencl.
Debugging is very painful for GPu kernels. Currently I have loop over number of
threads in my local machine.
Aseem
___
PyOpenCL mailing list
I think this is a good, simple & platform independent approach. There are
very few options when using pyopencl on nvidia GPUs.
Aseem
On Fri, Mar 9, 2018 at 10:09 PM, Karl Czajkowski wrote:
> On Mar 09, aseem hegshetye modulated:
> > Hi,
> > Is there a way to debug and/or
Hi,
I get this error when i run following code:
*import pyopencl as cl*
*cl.get_platforms()*
I tried following command in ubuntu terminal to fix it:
* sudo echo e- 'echo libnvidia-opencl.so.1' >>
/etc/OpenCL/vendors/nvidia.icd*
and
*echo libnvidia-opencl.so.1 >> /etc/OpenCL/vendors/nvidia.icd
I was able to log in as root user by doing "sudo -i" after ssh to the aws
ec2 instance.
but now the command *echo libnvidia-opencl.so.1 >>
/etc/OpenCL/vendors/nvidia.icd* says
"-bash: /etc/OpenCL/vendors/nvidia.icd: No such file or directory"
On Sun, Mar 11, 2018 a
tforms()
clGetPlatformIDs failed:
On Sun, Mar 11, 2018 at 4:58 PM, aseem hegshetye
wrote:
> I was able to log in as root user by doing "sudo -i" after ssh to the aws
> ec2 instance.
>
> but now the command *echo libnvidia-opencl.so.1 >>
> /etc/OpenCL/vendors/nvidia.icd* say
Hi,
What does following pyopencl error mean :
clEnqueueNDRangeKernel failed: INVALID_VALUE
How to solve it.
Aseem
___
PyOpenCL mailing list
PyOpenCL@tiker.net
https://lists.tiker.net/listinfo/pyopencl
Hi,
How to declare a custom function and call it inside a c kernel that runs on GPU
using pyopencl.
Thanks
Aseem
___
PyOpenCL mailing list
PyOpenCL@tiker.net
https://lists.tiker.net/listinfo/pyopencl
my Pyopencl code works perfect when ran on CPU by looping on thread_id. but
when ran on gpu on aws p2.xlarge ubuntu, an array who is specifically
initialized with 0's returns -1 instead of 0's. this arrays is of type
ctypes.c_int.
is there some other datatype for c int that I should be using.
thank
Hi,
I have a kernel as follows:
if (condition is true):# is always true
x=array[0,0];
if (cond2 is true):
x[i]++;
global_x[i]=x[i];
This gives me negative values in global_x. But when i declare x outside the
condition and copy x to global_x outside the if condition universally,
Hi ,
My pyopencl kernel just stops responding for specific data.
Code runs smoothly if data has total 30296 transactions. But becomes non
responsive when data has 30297 trans.
If I take that one extra trans and run the code separately on that trans , it
works fine.
Code can run on more than 40
Hi,
Does GPU speed exponentially drop as number of threads increase beyond a
certain number?. I used to allocate number of threads= number of
transactions in data under consideration.
For Tesla K80 I see exponential drop in speed above 30290 Threads.
If true, is it a best practice to keep number of
CS-2016-143.pdf
> Do you use atomic operations? Or maybe you have too many thread fences?
> I have no problem starting many threads: the number of threads alone is
> not the issues.
>
> Cheers,
> Sven
>
>
> Op 6-6-2018 om 8:37 schreef aseem hegshetye:
>
> Hi,
The findings below are considering I already have a 20 million * 57 bits int
array in the GPu.
> On Jun 6, 2018, at 3:05 AM, aseem hegshetye wrote:
>
> Hi,
> I did some testing with number of threads. I changed number of threads and
> recorded the time in seconds it took f
21 matches
Mail list logo
