# \#cudnn

**URL:** https://discourse.julialang.org/tag/cudnn/472.md

[Latest](https://discourse.julialang.org/latest.md) · [Categories](https://discourse.julialang.org/categories.md) · [Tags](https://discourse.julialang.org/tags.md)

---

## [Convolutions does not work on CUDA](https://discourse.julialang.org/t/convolutions-does-not-work-on-cuda/138703)

<div class="topic-metadata">

**Author:** [@Matej\_Zorek](https://discourse.julialang.org/u/Matej_Zorek)\
**Replies:** 0\
**Last updated:** [August 9, 2026, 12:49pm UTC](https://discourse.julialang.org/t/convolutions-does-not-work-on-cuda/138703 "2026-08-09T12:49:18Z")

</div>

Hi, I have a problem with Convolutional layers on gpu. It simply does not work. I have installed CUDA and cuDNN. GPU works for “simple” layers like Dense, LayerNorm etc, but when i try evaluate Conv it gives me an error…

---

## [Help using cuDNN in Julia](https://discourse.julialang.org/t/help-using-cudnn-in-julia/126232)

<div class="topic-metadata">

**Author:** [@notGiGi](https://discourse.julialang.org/u/notGiGi)\
**Replies:** 1\
**Last updated:** [February 25, 2025, 5:02pm UTC](https://discourse.julialang.org/t/help-using-cudnn-in-julia/126232 "2025-02-25T17:02:52Z")

</div>

Hello, everyone. I’m trying to use cuDNN in Julia, but so far it’s not working as expected. I’m very confused. It is supposed to be included in CUDA.jl so that I can use using CUDA.CUDNN, but when I do that, I get the er…

---

## [cuDNN, julia-1.10 and linux](https://discourse.julialang.org/t/cudnn-julia-1-10-and-linux/106658)

<div class="topic-metadata">

**Author:** [@quietlight](https://discourse.julialang.org/u/quietlight)\
**Replies:** 10\
**Last updated:** [January 28, 2024, 9:34pm UTC](https://discourse.julialang.org/t/cudnn-julia-1-10-and-linux/106658 "2024-01-28T21:34:28Z")

</div>

I have trained a Resnet18 using Flux on julia-1.10 rc1, actually I have trained multiple models with no trouble at all, but this week when I try to predict using my model I get a compilation error for my private package: …

---

## [Inconsitent gradients of BachNorm on GPU in testmode](https://discourse.julialang.org/t/inconsitent-gradients-of-bachnorm-on-gpu-in-testmode/105189)

<div class="topic-metadata">

**Author:** [@Phypix](https://discourse.julialang.org/u/Phypix)\
**Replies:** 0\
**Last updated:** [October 19, 2023, 2:49pm UTC](https://discourse.julialang.org/t/inconsitent-gradients-of-bachnorm-on-gpu-in-testmode/105189 "2023-10-19T14:49:11Z")

</div>

I encountered an issue with the gradient of the BatchNorm on the GPU in testmode. If the shape of the input array is either 2, 4 or 5, the gradients will differ from the result of any other sized array. The results on t…

---

## [Using real NCHW order when using cuDNN.jl](https://discourse.julialang.org/t/using-real-nchw-order-when-using-cudnn-jl/100842)

<div class="topic-metadata">

**Author:** [@Jonas208](https://discourse.julialang.org/u/Jonas208)\
**Replies:** 4\
**Last updated:** [June 30, 2023, 9:39pm UTC](https://discourse.julialang.org/t/using-real-nchw-order-when-using-cudnn-jl/100842 "2023-06-30T21:39:31Z")

</div>

I am trying to use cuDNN.jl for GPU accelerated convolution. I used the function cuDNN.cudnnConvolutionForward. The function has the keyword argument format that specifies the order of dimensions (format=cuDNN.CUDNN\_TEN…

---

## [Flux: reproducibility of GPU experiments](https://discourse.julialang.org/t/flux-reproducibility-of-gpu-experiments/62092)

<div class="topic-metadata">

**Author:** [@Rasmus\_Hoier](https://discourse.julialang.org/u/Rasmus_Hoier)\
**Replies:** 1\
**Last updated:** [May 30, 2021, 5:41pm UTC](https://discourse.julialang.org/t/flux-reproducibility-of-gpu-experiments/62092 "2021-05-30T17:41:00Z")

</div>

When I run the Flux model-zoo’s conv\_mnist.jl example there is an argument “seed” which should assure reproducibility (when larger than 0 ). This seems to only work when performing experiments on CPU, whereas GPU experim…

---

## [What is the fastest dimension ordering for convolutions?](https://discourse.julialang.org/t/what-is-the-fastest-dimension-ordering-for-convolutions/46182)

<div class="topic-metadata">

**Author:** [@maxfreu](https://discourse.julialang.org/u/maxfreu)\
**Replies:** 1\
**Last updated:** [January 25, 2021, 12:00pm UTC](https://discourse.julialang.org/t/what-is-the-fastest-dimension-ordering-for-convolutions/46182 "2021-01-25T12:00:57Z")

</div>

Hi! I am wondering what the fastest dimension ordering for convolution operations on the GPU using CUDNN currently is. Flux uses WHCN (column major), pytorch uses NCHW (row major) which is exactly the same. So to say, t…