https://github.com/rudojaksa/tfnohup2dat

extract data from the tensorflow (nohup) log
https://github.com/rudojaksa/tfnohup2dat

tensorflow

Last synced: about 2 months ago
JSON representation

extract data from the tensorflow (nohup) log

• Host: GitHub
• URL: https://github.com/rudojaksa/tfnohup2dat
• Owner: rudojaksa
• Created: 2020-03-17T22:41:26.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2020-03-17T22:57:09.000Z (over 6 years ago)
• Last Synced: 2025-02-22T21:14:06.433Z (over 1 year ago)
• Topics: tensorflow
• Language: Perl
• Homepage:
• Size: 15.6 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
### NAME

tfnohup2dat - extract data from the tensorflow nohup log

### USAGE

    nohup python tf-script.py &

    tfnohup2dat [OPTIONS] nohup.out > loss.dat

    tfnohup2dat [OPTIONS] -a nohup.out

    tf-script.py | tfnohup2dat

### DESCRIPTION

It extracts the sample number, loss, accuracy, etc. from the TensorFlow

standard progress report, as stored by nohup.

Compressed input files file.gz, file.lz, or .zst are also accepted.

### OPTIONS

     -h  This help.

     -v  Verbose messaging.

     -a  Automatic output file(s) file.dat (file-N.dat).  Multiple-files

         output for the multi-run nohup.out is possible in this mode.

     -r  Print raw input with non-ASCII characters removed.

### VERSION

tfnohup2dat-0.2 (c) R.Jaksa 2019,2020 GPLv3