denizyuret
denizyuret
Dear @manuEbg: The libknet8 library is only needed if you want to use the GPU. As far as I know this is not possible with the Macs at the moment...
@maleadt: we seem to have a problem in Knet where the installation gets stuck on a new architecture with no gpu support because it cannot locate the libknet8.so artifact. In...
@manuEbg can you try: ``` const libknet8 = Libdl.find_library(["libknet8"], try [artifact"libknet8"]; catch ex; []; end) ```
As a temporary workaround until I fix this, I would recommend: ``` KnetArray(repeat(CuArray(a),16,16))) ``` Or use `CuArray` as your array type, might be a bit slower but supports more operations.
Looking at the error message more carefully, it seems to be trying to find the gradient of uncat wrt its 4'th argument. The signature for uncat is: `uncat(dy, argn, dims,...
First, mixing of the old grad interface (i.e. grad(f)) and the new grad interface (grad(result, param)) is not well tested and part of the problem seems to be mixing the...
Dear Barışcan, could you try the following: 1. Try your model without the batchnorm operations. 2. Try your model with Array and/or CuArray instead of KnetArray for array type. 3....
This seems to be a CUDA.jl version issue. You may not have the right CUDA libraries installed. Please try to install and test CUDA.jl first. CUDA.versioninfo() should give you all...
I cannot replicate this error using Julia 1.6.2, CUDA 3.4.2 and Knet 1.4.8. On Mon, Aug 23, 2021 at 7:17 PM Mert Ceylan ***@***.***> wrote: > I am getting a...
TagBot fix