darknet/scripts/log_parser
Puneet Kohli cdd1cb0e8c Change matplotlib backend to 'agg' when 'show_plot' is false. This allows the script to be run from command line without any errors (for ex, on a remote machine) 2018-05-06 16:45:26 -05:00
..
log_parser.py Change matplotlib backend to 'agg' when 'show_plot' is false. This allows the script to be run from command line without any errors (for ex, on a remote machine) 2018-05-06 16:45:26 -05:00
plot.jpg parsing training log and plot 2018-05-06 22:11:49 +08:00
readme.md Update readme.md 2018-05-06 22:16:31 +08:00
run_log_parser_windows.cmd Minor fix - what pip-packages are required for Python scripts 2018-05-06 21:51:31 +03:00
test.log parsing training log and plot 2018-05-06 22:11:49 +08:00

readme.md

parsing training log and plot

Requirement

  1. ** matplotlib needed **

Usage

  1. --source-dir the directory of training log files
  2. --save-dir the directory to save loss curve, image and csv file
  3. --log-file log file name to be parsed
  4. --csv-file csv file name to save loss data, default it's same with training log file name
  5. --show whether to show after finished parsing, default False

python log_parser.py --source-dir ./ --save-dir ./ --log-file test.log --show true

plot