The codes on Caffe for ECCV 2018 paper Dividing and Aggregating Network for Multi-view Action Recognition
5
stars
0
forks
source link
Check failed: error == cudaSuccess (2 vs. 0) out of memory *** with just 6 videos #3
Closed
AntonioMarsella closed 5 years ago
Hi, I tried to run the code on just 6 videos of ntu dataset and I got the following:
Check failed: error == cudaSuccess (2 vs. 0) out of memory Check failure stack trace: @ 0x7f45615bf5cd google::LogMessage::Fail() @ 0x7f45615c1433 google::LogMessage::SendToLog() @ 0x7f45615bf15b google::LogMessage::Flush() @ 0x7f45615c1e1e google::LogMessageFatal::~LogMessageFatal() @ 0x7f45619019f2 caffe::SyncedMemory::mutable_gpu_data() @ 0x7f4561ae2fc8 caffe::CuDNNConvolutionLayer<>::Forward_gpu() @ 0x7f4561a756c2 caffe::Net<>::ForwardFromTo() @ 0x7f4561a758b7 caffe::Net<>::ForwardPrefilled() @ 0x7f4561abac0a caffe::Solver<>::Test() @ 0x7f4561abb726 caffe::Solver<>::TestAll() @ 0x7f4561ac3649 caffe::Solver<>::Step() I0930 10:49:18.202900 5791 video_data_layer.cpp:179] Restarting data prefetching from start. @ 0x7f4561ac4194 caffe::Solver<>::Solve() @ 0x4095c5 train() @ 0x406fa0 main @ 0x7f45604c3830 __libc_start_main @ 0x4074d9 _start @ (nil) (unknown) Setting device 0
What is the problem?
Thank you