Skip to content
This repository has been archived by the owner on Oct 1, 2019. It is now read-only.

make runtest stuck #158

Closed
thomhert opened this issue Oct 19, 2016 · 11 comments
Closed

make runtest stuck #158

thomhert opened this issue Oct 19, 2016 · 11 comments

Comments

@thomhert
Copy link

thomhert commented Oct 19, 2016

Hi Du Tran,
I checked several closed issues for "make runtest" but still couldn't find similar problem.
After doing make all and make test without error, I did "make runtest" but always got stuck at test point below. At first I thought it was because having two GPUs, then I do "make runtest CUDA_VISIBLE_DEVICES=0" or export CUDA_VISIBLE_DEVICES=0 before doing make runtest (based on here and here). However, i still got stuck at the same point.
Do you have any idea why I experienced this issue? Thank you.

Here is the test point that I got stuck:
[----------] 5 tests from Deconvolution3DLayerTest/1, where TypeParam = double
[ RUN ] Deconvolution3DLayerTest/1.TestSetup
[ OK ] Deconvolution3DLayerTest/1.TestSetup (0 ms)
[ RUN ] Deconvolution3DLayerTest/1.TestCPUGradient

or
[----------] 5 tests from Deconvolution3DLayerTest/1, where TypeParam = double
[ RUN ] Deconvolution3DLayerTest/1.TestSetup
[ OK ] Deconvolution3DLayerTest/1.TestSetup (0 ms)
[ RUN ] Deconvolution3DLayerTest/1.TestGPUSimpleDeconvolution3D
[ OK ] Deconvolution3DLayerTest/1.TestGPUSimpleDeconvolution3D (1 ms)
[ RUN ] Deconvolution3DLayerTest/1.TestCPUSimpleDeconvolution3D
[ OK ] Deconvolution3DLayerTest/1.TestCPUSimpleDeconvolution3D (1 ms)
[ RUN ] Deconvolution3DLayerTest/1.TestGPUGradient


I managed to solve the problem.. Thank you anyway.

@dutran
Copy link
Contributor

dutran commented Oct 20, 2016

@thomhert have you got your issue solved?

@thomhert
Copy link
Author

@dutran Yes, I finally solved it. So far I could try examples/c3d_feature_extraction.
Well, I guess my problem was because multi GPUs. However, to make it sure, I will re-try make runtest with multi GPUs and comment here if problems come up.

@dutran
Copy link
Contributor

dutran commented Oct 20, 2016

glad to hear!

@weizhou-geek
Copy link

@thomhert ,I also met this problem. How did you solve it? Thanks!

@thomhert
Copy link
Author

@weizhou-geek I assumed my problem was because there are two GPUs on my PC.
To solve this problem simply export CUDA_VISIBLE_DEVICES=0 before doing make runtest.
Follow these steps:

  • make clean (cleaning up the previous one)
  • make all -j $(($(nproc) + 1))
  • make test -j $(($(nproc) + 1))
  • export CUDA_VISIBLE_DEVICES=0
  • make runtest -j $(($(nproc) + 1))
    it seems stuck when I met this test point: [ RUN ] Deconvolution3DLayerTest/1.TestGPUGradient, so i just left it. When I checked it out in the morning, all test were finished.
    Hope it helps.

@weizhou-geek
Copy link

@thomhert , Thanks for your guidance. I have tried your method, but still stuck at the following point:
[----------] 3 tests from ImageDataLayerTest/1, where TypeParam = double
[ RUN ] ImageDataLayerTest/1.TestResize
[ OK ] ImageDataLayerTest/1.TestResize (464 ms)
[ RUN ] ImageDataLayerTest/1.TestShuffle
[ OK ] ImageDataLayerTest/1.TestShuffle (4600 ms)
[ RUN ] ImageDataLayerTest/1.TestRead
[ OK ] ImageDataLayerTest/1.TestRead (4477 ms)
[----------] 3 tests from ImageDataLayerTest/1 (9541 ms total)

[----------] Global test environment tear-down
[==========] 417 tests from 74 test cases ran. (24272972 ms total)
[ PASSED ] 417 tests.

YOU HAVE 2 DISABLED TESTS

@thomhert
Copy link
Author

@weizhou-geek what did you mean by stuck here? Because based on your comment, you passed 417 tests which means C3D is installed properly.

@weizhou-geek
Copy link

@thomhert But there still have two disabled tests.

@thomhert
Copy link
Author

thomhert commented Oct 21, 2016

@weizhou-geek just ignore it, your installation is done. Follow C3D guide to test your installation.
you can also read about disable test here.

@weizhou-geek
Copy link

@thomhert Thanks! Have you finished the example? I follow the C3D user guide and after feature extraction and volume mean creation, I use the command "sh train_ucf101.sh" but the procedures has been aborted:
I1021 19:19:25.417078 22871 train_net.cpp:26] Starting Optimization
I1021 19:19:25.417163 22871 solver.cpp:41] Creating training net.
I1021 19:19:25.417635 22871 net.cpp:76] Creating Layer data
I1021 19:19:25.417650 22871 net.cpp:112] data -> data
I1021 19:19:25.417671 22871 net.cpp:112] data -> label
I1021 19:19:25.417687 22871 video_data_layer.cpp:283] Opening file ../c3d_finetuning/train_01.lst
I1021 19:19:25.461426 22871 video_data_layer.cpp:312] Shuffling data
I1021 19:19:25.779758 22871 video_data_layer.cpp:317] A total of 107258 video chunks.
I1021 19:19:25.779816 22871 video_data_layer.cpp:344] read video from /data/users/trandu/datasets/ucf101/frm/JumpingJack/v_JumpingJack_g25_c05/
F1021 19:19:25.779863 22871 video_data_layer.cpp:346] Check failed: ReadImageSequenceToVolumeDatum(file_list_[id].c_str(), start_frm_list_[id], label_list_[id], new_length, new_height, new_width, sampling_rate, &datum)
*** Check failure stack trace: ***
@ 0x7f80fa829daa (unknown)
@ 0x7f80fa829ce4 (unknown)
@ 0x7f80fa8296e6 (unknown)
@ 0x7f80fa82c687 (unknown)
@ 0x4b68e3 caffe::VideoDataLayer<>::SetUp()
@ 0x458d4e caffe::Net<>::Init()
@ 0x45a005 caffe::Net<>::Net()
@ 0x433f6f caffe::Solver<>::Init()
@ 0x43767b caffe::Solver<>::Solver()
@ 0x40b726 main
@ 0x7f80f600dec5 (unknown)
@ 0x40e671 (unknown)
@ (nil) (unknown)
Aborted (core dumped)

@thomhert
Copy link
Author

thomhert commented Oct 21, 2016

@weizhou-geek you have the same issue to #132 #43 and i think you might not put your data path correctly.
see error here --> "read video from /data/users/trandu/datasets/ucf101/frm/JumpingJack/v_JumpingJack_g25_c05/" this is your wrong point.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants