Questions tagged [gpu]

Questions on using Mathematica functionality to take advantage of a Graphics Processing Unit, a special stream processor designed for highly parallel computer graphics operations.

Filter by
Sorted by
Tagged with
3 votes
0 answers
42 views

Can someone recommend workflow for debugging RemoteBatchSubmit jobs on AWS

I am trying to troubleshoot using an external service (AWS: Amazon Web Service) for NetTrain with a GPU. My RemoteBatchSubmit has been stuck at "Runable" for about 10 hours as described in ...
  • 2,881
5 votes
2 answers
274 views

Optimizing sequential computation

The following question came up in reading group discussion of paper -- imagine taking two vectors, multiplying each by a random matrix, then taking tanh pointwise. If you repeat this process forever, ...
0 votes
0 answers
98 views

OpenCL with Mathematica 12

I got CUDA to work but no OpenCL: "OpenCLLink is not supported on this system." ...
1 vote
0 answers
86 views

Unknown Error in NetGraph with RTX 3090 GPU

I am running NetGraph on my RTX 3090 GPU. The Net itself loads properly in my GPU, having an approximate size of 9GB (the 3090 has 24GB of memory). However, once the net actually starts NetGraph ...
0 votes
1 answer
53 views

Does the Solve function as applied to simultaneous equations benefit from multi-node / GPU parallel computing?

I use the Solve function to find solutions for systems of simultaneous equations with dozens or hundreds of equations and variables. This is successful though very slow (weeks) on a conventional ...
  • 31
2 votes
1 answer
131 views

Anyone has any succes on GPU NetTrain with RemoteBatchSubmit on AWS EC2?

Since 12.3, the RemoteBatchSubmit to EC2 does not work anymore. First, it features the same bug as Version 12.3.1 Problems with NetTrain, where the MXNetResources paclet is 12.4 instead of 12.3. ...
2 votes
0 answers
82 views

Solving differential equations with GPU?

I have a complicated differential equation which is taking a lot of time to solve, if I had more cores it would be easier, but my computer is limited... I was wondering: is it possible to solve ...
  • 6,214
4 votes
1 answer
193 views

How can I free GPU memory allocated with NetTrain

When I train a neural network with NetTrain[..., TargetDevice->"GPU"], wolfram's neural network code automatically allocates memory on my GPU for my ...
1 vote
2 answers
176 views

Optimizing the calculation of a global maximum

...
2 votes
1 answer
117 views

Neural network evaluate on multiple gpus simultaneously

For instance, I have 4 gpus, now I want to calculate neural network on each gpu simultaneously, or make use of all of my gpus to calculate the data. The code might like ...
2 votes
0 answers
95 views

How to use _Complex types in CUDAFunctionLoad?

The documentation for CUDAFunctionLoad specifies various base datatypes you can use for your function's arguments, including ...
3 votes
0 answers
57 views

Unable to uninstall an older CUDAResource paclet

I am trying to uninstall an older version of CUDAResource but cannot do so. Any ideas please? Is there a manual way to do so? ...
3 votes
0 answers
120 views

Classify/NetTrain run indefinitely without output when using GPU

I am experimenting with Mathematica's Classify/NetTrain functions for the first time. When I don't set a TargetDevice option, Mathematica defaults to the CPU and everything works as expected. When I ...
  • 101
2 votes
2 answers
277 views

Does NetTrain Support Nvidia 30xx series GPU (RTX 3080, 3090 etc)?

Recently, I want to buy some 30xx series GPUs which are used to do deep learning jobs, so does anyone know whether Mathematica's NetTrain supports 30xx series GPUs?
4 votes
1 answer
143 views

Mathematica GPU occupancy is too high when the current page has high quality images

When Mathematica's current page contains a high-quality picture, the GPU occupancy rate will increase and the current page delay of Mathematica will increase( the software is slow to respond to my ...
  • 189
11 votes
0 answers
217 views

Tutorial example NetTrain fails miserably with a GPU

I have been experimenting with GPU NetTrain on AWS now when Mma 12.2 supports remote batch jobs (I don't have an Nvidia GPU to try these things out otherwise). In ...
  • 15.6k
7 votes
0 answers
165 views

Extremely Long GPU Initialization Times on Mathematica 12.2 and RTX3090 [closed]

I'm having trouble getting my new GPU system to work with Mathematica 12.2. I have two RTX3090 cards in my setup. My NVIDIA drivers are version 460.89 (the latest). I have installed CUDAToolkit 11.0 ...
15 votes
0 answers
145 views

How to use user-provided CUDA toolkits?

In the revision history, Mathematica v12.1 claims to have "support for user-provided CUDA Toolkits". But other than this line in the revision history, I cannot find any other reference to ...
  • 1,146
3 votes
1 answer
501 views

CUDA usage in "everyday life"

I have checked documentation about CUDA usage in Mathematica and it seems that it is quite few applications of CUDA. Are the any additional packages/functions/tutorials/documentations for CUDA in ...
2 votes
2 answers
175 views

Why is Graphics3D lighting different when rotating? [duplicate]

Mathematica v12.1 on macOS Catalina 10.15.5. I'm getting weird Lighting behavior in Graphics3D in the sense that the rendered ...
  • 199
3 votes
1 answer
4k views

GPU support on latest MacBook Pro, especially for deep learning

I'm thinking about getting some new top of the line computer hardware, specifically a MacBook Pro from Apple. These have AMD Radeon Pro 5500M GPUs, which while a bit slower than something like an ...
  • 187
12 votes
4 answers
572 views

CUDA on Tesla V100 card

Context I am trying to run CUDALink with a Centos 6.10 Mathematica 12.0 [cuda 10.1]. Host has NVIDIA Corporation GV100GL [Tesla V100 PCIe 32GB] (rev a1) card ...
  • 22k
1 vote
0 answers
48 views

NetTrain with Quadro

I have an Nvidia Quadro P620/PCIe/SSE2 graphics card on my linux box (Ubuntu 18.04): <<CUDALink` CUDAQ[] (*True*) CUDADriverVersion[] (* 390.116.0 *) But ...
  • 3,551
1 vote
0 answers
92 views

Re-train a neural network discarding its memory?

...
0 votes
0 answers
414 views

GPU Computation AMD

I'm new to this forum even though it has helped me a lot over the last few months. I'm working on hypergeometric functions and want to use my GPU as support. I have a Radeon R9 390, but I can't get it ...
4 votes
1 answer
219 views

Using GPU Computation in LearnDistribution

LearnDistribution has an option for RealNVP, which is directly training a neural network for the distribution. It also has an option for TargetDevice, however, it appears to do nothing. Take the ...
  • 3,584
0 votes
2 answers
274 views

CUDA Trouble, MMA12 Quadro RTX 4000

Started this unfortunately as an answer in this question on GPUs under Mma12: previous GPU question, while it should have gone into a new question. I will try to answer questions already posted on the ...
3 votes
2 answers
290 views

Does Mathematica 12 support GT 730 CUDA?

I am getting the following error when I set the target device as GPU: NetTrain::trgdevegpu: TargetDevice -> GPU could not be used; a supported NVIDIA GPU was not found. If you are using an ...
  • 173
1 vote
0 answers
106 views

Is Wolfram Cloud going to support GPU training? [duplicate]

Is Wolfram Cloud going to support GPU training? Is there any other possibility for training a neural net (in WL) on a cloud like AWS, Floydhub, or so? I think this concerns lots of people.
3 votes
1 answer
971 views

CUDA on Manjaro Linux (Arch)

I have been struggling getting CUDA to work on Mathematica 11.3 and Manjaro Linux (kernel 4.19.16-1-MANJARO). Trying some commands I get the following: ...
  • 927
2 votes
0 answers
78 views

NetTrain with GPU crashed

I tried to run NetTrain with TargetDevice->"GPU", and Mathematica crashed. I got the following message: When I executed ...
  • 121
1 vote
0 answers
124 views

CUDACCompilers[] returns an empty list

I have just installed Mathematica 11.3.0.0 on Windows 10 (Professional), with an NVIDIA GT 630M GPU on my Dell Inspiron. I have Visual Studio 2010 Express C++ Compiler installed on my system. I am ...
1 vote
0 answers
67 views

Dimension of Grid in CUDAFunctionLoad for Matrix Multiplication

I want to do nonsquare matrix multiplication using CUDA The input is two matrices: a and b ...
  • 11
0 votes
0 answers
136 views

CUDAQ::internal: CUDALink experienced an internal error

I have Mathematica 11.0.0 installed on Windows 10, with an NVIDIA GT 630M GPU on my Dell Inspiron. I am trying to run the CUDA examples in the Help files. All of the setup functions for CUDALink work ...
5 votes
2 answers
315 views

OpenCL compute large matrices

I am trying to compute a big matrix $C$, of size $M \times N$, using the elements of the input matrices $A$ and $B$. The matrix $A$ is of size $M\times4$ and the matrix $B$ is of the size $2\times N$. ...
  • 341
2 votes
0 answers
121 views

Whether Parallel computing or Gpu Computing can be performed for mathematica built-in functions? [closed]

I want to know how to run my mathematica code using GPU or parallel computing without any special coding besides using the original Mathematica code.
5 votes
0 answers
284 views

Why doesn't NetTrain use GPU as TargetDevice?

I have a NetTrain call with the following parameters: tNet = NetTrain[inVecs -> outVecs, TargetDevice -> "GPU"]; But ...
  • 4,871
2 votes
1 answer
72 views

Target system specification is not available for NVIDIA CUDA Compiler installation

I Installed Mathematica correctly and it indeed found my GPU. My case is exactly the same as : the same issue link However, it is not answered. Does anyone know the reason? My case: ...
1 vote
1 answer
430 views

Sparse matrices dot multiplication

I have two huge sparse matrices, and I am interested in dot multiplication them. But I have a memory issue so a notebook is a shutdown. I am using the 11.3 version. Any suggestion on how to avoid ...
7 votes
1 answer
1k views

CUDA for linear equations

TLDR; Is there a way to solve linear equations of a sparse matrix (discretized laplace operator) efficiently using CUDALink in Mathematica? I didn't find a CUDALinearSolve or CUDAMatrixInverse or ...
1 vote
0 answers
49 views

How to get command line for OpenCLFunctionLoad?

Is there a way how to get the command used to compile my OpenCL kernel, when I use OpenCLFunctionLoad[]? When I use ...
  • 171
2 votes
0 answers
1k views

Does Mathematica Rely More on CPU or GPU

I am surprised no one asked this before. I am getting a new laptop and I will be mainly using Mathematica more than anything else on it. When selecting a laptop, should I put more emphasis on CPU or ...
  • 165
1 vote
1 answer
236 views

Using library code (specifically curand) with CUDALink

...
  • 13.6k
10 votes
1 answer
254 views

Can one use NetTrain with two Nvidia GPUs?

I am thinking of buying another GPU to accelerate my training. Will I be able to call NetTrain with both GPUs ("TargetDevice"-> {"GPU", All})? Or would it be ...
  • 1,525
3 votes
1 answer
223 views

CUDAQ[] returns false in Mathematica 10.0.2.0, Mac OS 10.13.2, CUDA Driver Version 387.99

CUDAQ[] returns False on my Mac OS 10.13.2. I typed Needs["CUDALink`"] and it returns no error. ...
3 votes
1 answer
235 views

Why the GPU will cost more time when train the net

I'm in Windows 10 $Version 11.2.0 for Microsoft Windows (64-bit) (September 11, 2017) ...
  • 24.6k
11 votes
1 answer
799 views

How to use an external GPU to train neural networks in Mathematica?

Taking steps in the uncharted territories of eGPUs, I have installed an NVIDIA GeForce Titan Xp GPU on an eGPU enclosure (Mantiz MZ-02) and have connected it to my laptop (HP Spectre x360 - 15-bl075nr ...
  • 1,813
7 votes
1 answer
530 views

Why I can only train some NetModel with GPU

I can train NetModel's models with CPU. I'm on Windows 10 with Mathematica 11.2 GPU is GTX1080 with 8G memory. My datasets is 1000 pictures with 224*224. I can train some NetModel like ...
  • 8,277
4 votes
0 answers
313 views

NetTrain throws internal error with large images as input (GPU)

I am training a convolutional Network with a dataset consisting of Images of the same size on my GPU (200x100 pixels, GTX 1080Ti). The images are previously resized with ...
  • 1,525
2 votes
1 answer
393 views

CUDALink and eGPU for Mac?

I have: MacBook Pro mid 2014 13" (i.e. just an onboard Intel graphics card) NVIDIA GeForce Titan X Akitio Node Thunderbolt 2 Cable Thunderbolt 2 to Thunderbolt 3 Adaptor I have installed the NVIDIA ...
  • 5,262