search

layumi / Pedestrian_Alignment

TCSVT2018 Pedestrian Alignment Network for Large-scale Person Re-identification
https://ieeexplore.ieee.org/document/8481710
MIT License
235 stars 66 forks source link

Fail in compilation #5

Closed xueyuxiaolang closed 7 years ago

xueyuxiaolang commented 7 years ago

when I uncomment and modify some lines in gpu_compile.m and run it in Matlab. I find error and I don't know how to solve this problem. `

gpu_compile Error using vl_compilenn Too many output arguments.

Error in gpu_compile (line 5) vl_compilenn('enableGpu', true,...`

ps：I have run vl_testnn('gpu', true) and show ` Failure Summary:

 Name                                                                    Failed  Incomplete  Reason(s)
=======================================================================================================
 nnmnist[dataType=single,device=gpu]/valErrorRate(networkType=dagnn)       X         X       Errored.
-------------------------------------------------------------------------------------------------------
 nnmnist[dataType=single,device=gpu]/valErrorRate(networkType=simplenn)    X         X       Errored.

result =

1×1673 TestResult array with properties:

Name
Passed
Failed
Incomplete
Duration
Details

Totals: 1671 Passed, 2 Failed, 2 Incomplete. 482.2786 seconds testing time.`

ckmessi commented 6 years ago

@xueyuxiaolang Hi, I encountered the same problem as you said. Have you resolved it and how to resolve it? Thanks.

layumi commented 6 years ago

No worries. I used vl_testnn and it also output error. More importantly, can you run the code without error? Sometimes gpu/ readjpg will go wrong.

ckmessi commented 6 years ago

@layumi Thank you for your reply. Actually, it's my first time to code with MatConvnet, so it's a little confusing at first. Now I am waiting for finishing downloading the data "imagenet-resnet-50-dag.mat", which is so slow from the vlfeat site. After that, I'll try the functional code. Thank your for your share.