Community Profile

photo

Joss Knight

MathWorks

Last seen: ein Tag vor |  Aktiv seit 2013

Although I cannot be contacted directly, if you would like to ask me a question all you have to do is mention "GPU" somewhere in your MATLAB Answers question.

Statistiken

  • Guiding Light
  • 36 Month Streak
  • Knowledgeable Level 5
  • Revival Level 2
  • First Answer

Abzeichen anzeigen

Content Feed

Anzeigen nach

Beantwortet
arrayfun vs loops again
I wish it were a cleverer answer, but I'm afraid that it's simply that MATLAB has been heavily optimized for |for| loops over th...

ein Tag vor | 0

Beantwortet
Extreme memory usage when using gather
As others have worked out, it looks like the issue is the indexed assignment into u: u(TERRAIN:end,m) = abs(gather(head(ffield,...

15 Tage vor | 0

| akzeptiert

Beantwortet
How can i fix the error mismatch format when convert multiplication layer from LayerGraph to dlnetwork in prune and quantize network example
For one solution, replace the fully connected layers with convolution layers with filters the size of the input and num filters ...

18 Tage vor | 0

Beantwortet
How can i fix the error mismatch format when convert multiplication layer from LayerGraph to dlnetwork in prune and quantize network example
That's an annoying limitation which is hopefully fixed in current releases. Is there any particular reason why you can't use |mu...

18 Tage vor | 0

Beantwortet
Arrayfun/gpuArray CUDA kernel need to be able to remember previous steps
This is a bit too vague to answer. Without indexing, how can each subproblem retrieve its subset of the data? If you just mean i...

18 Tage vor | 0

Beantwortet
Loop through list of vectors (ie, the rows of a matrix), applying same "simple" function to each one, on GPU? Should I use arrayfun somehow?
If your calculation is truly sequential then by definition you cannot parallelize along the sequence. But it sounds like you def...

18 Tage vor | 0

Beantwortet
Parallel Computing in C when using Matlab Coder (parpool and gpuArray)
GPU Coder will generate CUDA code for you. It can even automatically vectorize |for| loops. For a multithreaded parallel for loo...

3 Monate vor | 1

Beantwortet
Can parfor run a series of GPU programs simultaneously？
It looks like you just have a bug in your CUDAKernel implementation, probably accessing unallocated memory. This is putting the ...

3 Monate vor | 0

Beantwortet
error of GPU, net = trainNetwork(datastore, lgraph, options);
You are using your display GPU for computation and it does not have enough capacity. Try disabling all hardware acceleration for...

3 Monate vor | 1

| akzeptiert

Beantwortet
How to train a sequence to classification network on GPU
This performance discrepancy is normal. Small sequence networks often cannot benefit from GPU parallelism, especially if they us...

3 Monate vor | 0

Beantwortet
gpuArray large sparse arrays. Error codes: "CUSPARSE_INTERNAL_ERROR" / "UNKNOWN_ERROR"
Hi Joseph. It's hard to be definitive. There were some problems with cusparse and also Windows drivers when supporting the newes...

3 Monate vor | 0

Beantwortet
NVIDIA A2 performance in Matlab R2023b is lousy
The A2's spec says its double precision performance is 140 GFLOPS vs the V100's 7 TFLOPS, so this is pretty much expected. Even ...

6 Monate vor | 3

| akzeptiert

Beantwortet
Is there a utility like nvidia-smi within matlab to determine which gpus are in current use?
gpuDeviceTable is the utility for listing your devices and their properties. You can also see all your devices in the Parallel m...

6 Monate vor | 0

Beantwortet
Optimizing distance calculation between vectors and pixels
I feel like I haven't fully understood what you're after here, but |pdist2| is the function you're supposed to use to compute di...

7 Monate vor | 0

Beantwortet
The value of 'ValidationData' is invalid. Duplicate table variable name: 'input'. error during neural network training
augmentedImageDatastore returns a table so cannot be trivially combined. You should first transform it to convert it into a cell...

7 Monate vor | 0

| akzeptiert

Beantwortet
How to apply a 2D matrix input to a trainNetwork?
Consult the documentation here. Typically sequence data is passed in as a cell array. In each cell you would be passing one seq...

7 Monate vor | 0

Beantwortet
I would like to train a smallish network using cpu cores in parallel rather than gpu as they are slower.
setenv CUDA_VISIBLE_DEVICES -1 when you first start MATLAB, assuming you are running everything locally. However, as a general...

7 Monate vor | 0

Beantwortet
Faster three dimensional higher order interpolation?
Thanks for the request, it will help us prioritise future work. In the meantime, it is possible to write your own interpolation ...

8 Monate vor | 0

Beantwortet
How to solve "Unexpected error calling cuDNN: CUDNN_STATUS_EXECUTION_FAILED." error?
The Ada GPU architecture is not supported in R2018b. You need to upgrade MATLAB. You should have received a warning about this w...

8 Monate vor | 0

| akzeptiert

Beantwortet
Will self-written exe application run on GPU on other PC?
Yes, if your other PC has a supported GPU your application will run on it. Isn't MATLAB great?!

9 Monate vor | 0

| akzeptiert

Beantwortet
Error at linking stage using mexcuda
Relocatable device code needs to be linked by nvcc using -dlink before it can be linked to host code using the C linker, so you'...

9 Monate vor | 0

| akzeptiert

Beantwortet
OCR returns slightly different results on different machines
This is expected for any highly optimized code like this. Even for two Intel machines, the core count will affect how operations...

9 Monate vor | 1

Beantwortet
Will the MATLAB Answers community diminish/obsolete with the rise of AI-based chatbots?
You look like you are asking a question about how AI-assisted automation will change MATLAB Answers in the coming years and pote...

10 Monate vor | 3

Beantwortet
The matlab mexw64 file generated by mexcuda cannot be executed in the standalone app generated by matlab('parallel.gpu.GPUDeviceManager.selected' cannot be detected))
MATLAB Compiler's dependency analyzer cannot detect your dependency on PCT. Either add the product manually or call something ex...

10 Monate vor | 0

Beantwortet
Unable to perform assignment because the indices on the left side are not compatible with the size of the right side.
If you click on the line number in the editor next to where you create your function layer, you can put a breakpoint at the entr...

10 Monate vor | 0

| akzeptiert

Beantwortet
Using a "CUDAKernel" type object within a parfor loop
A CUDAKernel object cannot be serialized, as you've found, so you will need to construct it separately on each worker. However, ...

10 Monate vor | 2

| akzeptiert

Beantwortet
Performance drop on mobile RTX4080
The 4080 is a good 10x slower than the V100 in double precision so this doesn't surprise me - it is designed for workstation gra...

10 Monate vor | 0

Beantwortet
Using experiment manager on single GPU
It does depend on the balance of CPU and GPU work in your experiment, but as a general rule parallel execution will gain you not...

10 Monate vor | 0

Beantwortet
Memory issue with texture in mexCUDA compiled code
It looks like the syntax for your function |mxArrayToTexture_3D_float4| is incorrect. You are passing the pointer |cuArray| by v...

11 Monate vor | 0

Beantwortet
host compiler failed but others all passed with 'coder.checkGpuInstall('full')'
MSVC 2022 was not supported by the NVIDIA CUDA compiler in R2022b. You either need to install MSVC 2019 (or 2017) or upgrade MAT...

11 Monate vor | 0

Mehr laden