我正在用电筒运行程序,在终端中键入以下命令时出现错误。
th run.lua --dataset camvid --datapath ./data/CamVid/ --nThreads 3 --save ./models --imHeight 360 --imWidth 480 --maskHeight 360 --maskWidth 480 --batchSize 2 --model segnet --plot --learningRate 0.01 --learningRateDecaySteps 50 --maxepoch 3 --confusion all
错误是:
/usr/local/intel-17/openmpi-2_0_1/Torch/7/bin/luajit: ...el-17/openmpi-2_0_1/Torch/7/share/lua/5.1/trepl/init.lua:389: ...el-17/openmpi-2_0_1/Torch/7/share/lua/5.1/trepl/init.lua:389: ...-17/openmpi-2_0_1/Torch/7/share/lua/5.1/cutorch/init.lua:2: cannot load '/usr/local/intel-17/openmpi-2_0_1/Torch/7/lib/lua/5.1/libcutorch.so'
stack traceback:
[C]: in function 'error'
...el-17/openmpi-2_0_1/Torch/7/share/lua/5.1/trepl/init.lua:389: in function 'require'
run.lua:10: in main chunk
[C]: in function 'dofile'
...nmpi-2_0_1/Torch/7/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x00406560
我正在使用Torch / 7和cuda 7在ubuntu 14.04上运行它。您能否让我知道此错误是否又回到了cuda和torch的版本,还是与代码有关? 我真的很困惑如何解决这个问题。