Dependencies "4 skipped during auto due to previous errors"

General Usage Performance
1

When I follow the tutorial on :

When I instantiate in julia: (image attached):

image
image1108×252 20.1 KB

I obtain the following errors:

image
image2114×196 34.4 KB

Do someone knows how to fix the problem of these dependencies?

2

Did you run import Pkg; Pkg.precompile() or load the package as the message suggests?

3

I activate and then instantiate, after I instantiate the following error are shown:

image
image1254×570 81.4 KB

The image I attached in the main question is after running:

import Pkg; Pkg.precompile()

4

When I add the package “cuda” I also got the error:

image
image1252×592 117 KB

5

Dear, @ianshmean i got what you meaned: I already add manually the packages But I still have problems with two of them:
Flux
NormalFormAE.

adding manually the packages is not working. When I runed run

import Pkg; Pkg.precompile()

I do not understand where the error comes from. Find the error attached:

image
image2452×1608 454 KB

Do you got the same error?

6

Assuming that the checked-in manifest actually works with some julia version… The problem here seems to be that you’re trying to instantiate a manifest that was resolved in a different julia version, but unfortunately older versions of julia didn’t record which version that was, and the repo also doesn’t mention which julia version to use, so it’s just an informed guess that that’s what the issue is.

I recommend trying to update the environment rather than instantiate.

Also, when posting an issue like this, it’s helpful to:

7

Thanks it worked. Now I have another error. I think I will open a new topic for this new error.

shell> julia -i run/run_nf.jl
/usr/share/modules/init/bash: eval: line 43: syntax error near unexpected token (' /usr/share/modules/init/bash: eval: line 43: Pkg.precompile() _mlshdbg=‘’ ;;’
Activating project at ~/Documents/NormalFormAE
ERROR: LoadError: CUDA error: invalid device ordinal (code 101, ERROR_INVALID_DEVICE)
Stacktrace:
[1] throw_api_error(res::CUDA.cudaError_enum)
@ CUDA ~/.julia/packages/CUDA/Uurn4/lib/cudadrv/error.jl:91
[2] macro expansion
@ ~/.julia/packages/CUDA/Uurn4/lib/cudadrv/error.jl:101 [inlined]
[3] cuDeviceGet
@ ~/.julia/packages/CUDA/Uurn4/lib/utils/call.jl:26 [inlined]
[4] CuDevice(ordinal::Int64)
@ CUDA ~/.julia/packages/CUDA/Uurn4/lib/cudadrv/devices.jl:17
[5] top-level scope
@ ~/Documents/NormalFormAE/run/run_nf.jl:6
in expression starting at /home/munozdjp/Documents/NormalFormAE/run/run_nf.jl:6

8

I don’t expect it to be the cause of the error you’re seeing, but just FYI, you appear to be running julia from within the julia shell>. I recommend keeping it simple and just run julia in a plain terminal

9

Thanks it seems that my code had written

device!(CUDA.CuDevice(1))

Changing 1 by 0 fixed the problem. not sure yet why.

10

You probably only have a single GPU, while that function takes a 0-indexed number.