1. 09 Aug, 2022 1 commit
  2. 13 Jul, 2022 1 commit
  3. 16 Jun, 2022 1 commit
  4. 15 Jun, 2022 1 commit
  5. 14 Jun, 2022 1 commit
  6. 18 Jan, 2022 1 commit
  7. 03 Dec, 2021 1 commit
  8. 13 Nov, 2021 1 commit
  9. 25 Oct, 2021 1 commit
  10. 20 Oct, 2021 2 commits
  11. 07 Sep, 2021 1 commit
  12. 26 Jul, 2021 1 commit
  13. 17 Jun, 2021 1 commit
  14. 02 Jun, 2021 1 commit
  15. 28 May, 2021 1 commit
  16. 27 May, 2021 1 commit
  17. 10 Apr, 2021 1 commit
  18. 09 Apr, 2021 1 commit
  19. 30 Mar, 2021 1 commit
  20. 25 Mar, 2021 1 commit
  21. 22 Mar, 2021 1 commit
    • A. Unique TensorFlower's avatar
      Add a new option "recreate_iterator_for_each_eval" into StandardEvaluator. · 0b7674b9
      A. Unique TensorFlower authored
      In current implementation of StandardEvaluator, the iterator of evaluate dataset will go back (actually create a new iterator) to the beginning when evaluate() is called every time.
      
      In some case, the iterator need to go ahead forever. So add the new option "recreate_iterator_for_each_eval" to control this behavior.
      
      In some case, the iterator creation is time consuming. For example, careting an iterator of distributed dataset for multi worker in eager mode. So it's better to make a repeat dataset instead of recreating an iterator.
      
      If "recreate_iterator_for_each_eval" is True, then the iterator will go back; Otherwise, keep going ahead.
      
      PiperOrigin-RevId: 364356169
      0b7674b9
  22. 18 Mar, 2021 1 commit
  23. 13 Mar, 2021 1 commit
  24. 10 Mar, 2021 1 commit
  25. 05 Feb, 2021 1 commit
  26. 14 Jan, 2021 1 commit
  27. 13 Jan, 2021 1 commit
  28. 05 Jan, 2021 1 commit
  29. 29 Dec, 2020 1 commit
  30. 24 Dec, 2020 1 commit
  31. 11 Dec, 2020 1 commit
  32. 19 Nov, 2020 1 commit
  33. 17 Nov, 2020 1 commit
  34. 13 Nov, 2020 3 commits
  35. 10 Nov, 2020 1 commit
  36. 01 Nov, 2020 2 commits