deep learning - caffe parse log is broken since rc5 -


i have upgraded caffe/rc5 , figured out parse_log.sh not date anymore. there new way produce parsed log or bug?

produced output in log.train:

#iters seconds trainingloss learningrate 0    62.534432   62.5355s/100  0.0005 100  119.882181  57.3488s/100  0.0005 

corresponding log file

i0328 11:28:27.227087  9438 solver.cpp:398]     test net output #0: loss = 907.84 (* 1 = 907.84 loss) i0328 11:28:27.680057  9438 solver.cpp:219] iteration 0 (0 iter/s, 62.5355s/100 iters), loss = 895.13 i0328 11:28:27.680119  9438 solver.cpp:238]     train net output #0: loss = 860.911 (* 1 = 860.911 loss) i0328 11:28:27.680124  9438 sgd_solver.cpp:105] iteration 0, lr = 0.0005 i0328 11:29:25.027806  9438 solver.cpp:219] iteration 100 (1.74372 iter/s, 57.3488s/100 iters), loss = 56.9938 i0328 11:29:25.028008  9438 solver.cpp:238]     train net output #0: loss = 64.645 (* 1 = 64.645 loss) i0328 11:29:25.028014  9438 sgd_solver.cpp:105] iteration 100, lr = 0.0005 


Comments

Popular posts from this blog

javascript - Clear button on addentry page doesn't work -

c# - Selenium Authentication Popup preventing driver close or quit -

tensorflow when input_data MNIST_data , zlib.error: Error -3 while decompressing: invalid block type -