1. 17 Jun, 2022 1 commit
  2. 10 Jun, 2022 1 commit
  3. 03 Jun, 2022 1 commit
  4. 25 May, 2022 2 commits
  5. 28 Apr, 2022 1 commit
  6. 27 Apr, 2022 1 commit
  7. 26 Apr, 2022 1 commit
  8. 21 Apr, 2022 1 commit
  9. 07 Apr, 2022 1 commit
  10. 05 Apr, 2022 2 commits
  11. 29 Mar, 2022 1 commit
    • A. Unique TensorFlower's avatar
      Improved error message by indicating detected tf version, for easier... · 35176131
      A. Unique TensorFlower authored
      Improved error message by indicating detected tf version, for easier debugging. Eg, now reads "ValueError: hourglass_20 is not supported for tf version 1. See `model_builder.py` for features extractors compatible with different versions of Tensorflow".
      
      Before, the message was confusing because it might've indicated hourglass_20 was not supported (but it is... just not for tf version 1).
      
      PiperOrigin-RevId: 438053726
      35176131
  12. 28 Mar, 2022 2 commits
  13. 25 Mar, 2022 2 commits
  14. 15 Feb, 2022 1 commit
  15. 14 Feb, 2022 2 commits
  16. 05 Feb, 2022 1 commit
  17. 03 Feb, 2022 1 commit
  18. 02 Feb, 2022 1 commit
  19. 20 Jan, 2022 1 commit
  20. 11 Jan, 2022 3 commits
  21. 06 Jan, 2022 1 commit
  22. 20 Dec, 2021 1 commit
  23. 13 Dec, 2021 1 commit
  24. 11 Dec, 2021 1 commit
  25. 09 Dec, 2021 1 commit
  26. 07 Dec, 2021 1 commit
  27. 03 Dec, 2021 1 commit
  28. 23 Nov, 2021 1 commit
  29. 19 Nov, 2021 1 commit
  30. 17 Nov, 2021 1 commit
    • Scott Main's avatar
      Generalize the xml parsing to facilitate custom datasets. · 743c28aa
      Scott Main authored
      Instead of searching the annotation files based on class names in the filenames, read all the XML annotation files. For the cats/dogs use-case, just delete all the XML files that are not the breeds we care about.
      
      PiperOrigin-RevId: 410613750
      743c28aa
  31. 16 Nov, 2021 1 commit
  32. 04 Nov, 2021 2 commits