1. 22 11月, 2016 2 次提交
  2. 21 11月, 2016 1 次提交
  3. 20 11月, 2016 1 次提交
    • Y
      Fix several cpp issues · 20600e70
      Yu Yang 提交于
      * Different Type compare.
      * ostream << should pass a const object.
      * remove always true checks.
      20600e70
  4. 17 11月, 2016 1 次提交
  5. 13 11月, 2016 1 次提交
  6. 12 11月, 2016 1 次提交
  7. 11 11月, 2016 1 次提交
  8. 04 11月, 2016 1 次提交
  9. 18 10月, 2016 1 次提交
  10. 17 10月, 2016 2 次提交
    • E
      Fix sparse training for trainer_count=1 (#204) · 28bc05b1
      emailweixu 提交于
      * Fix sparse training for trainer_count=1
      
      For trainer_count=1, the gradient machine is NeuralNetwork, which does not create parameter buf for PARAMETER_GRADIENT for sparse update in Parameter::enableType. But gradient parameter buf is still used in SgdThreadUpdater.
      
      * Minor update to comment
      28bc05b1
    • Y
      Fix bug in yield dictionary in DataProvider. (#197) · 2f82d72e
      Yu Yang 提交于
      * Fix bug in yield dictionary in DataProvider.
      * Also make virtualenv work in Paddle.
      2f82d72e
  11. 10 10月, 2016 1 次提交
  12. 08 10月, 2016 1 次提交
  13. 29 9月, 2016 3 次提交
  14. 28 9月, 2016 1 次提交
  15. 26 9月, 2016 1 次提交
  16. 23 9月, 2016 1 次提交
  17. 20 9月, 2016 2 次提交
    • L
      comment LOG(info) in createPythonClass · 8b1c76af
      liaogang 提交于
      * it makes unit test failed.
      8b1c76af
    • Y
      Add min_pool_size, Add default value of should_shuffle (#70) · 90b9cba7
      Yu Yang 提交于
      * min_pool_size would be infinite by default.
        * add unittest for min_pool_size
      * Fix bug in can_over_batch_size
        * add unittest for can_over_batch_size
      * Add DEFINE_PROVIDER_EX
      * Add default value of should_shuffle
        * When training, the default value of should_shuffle is True.
        * When testing, the default value of should_shuffle is False.
        * User a set a provider should_shuffle or not by pass it to `@provider`
        * should_shuffle can handle a list of value, not just boolean
      * Add input order mapping by using name
        * Add unittest
      * Add check to check input format.
        * Default is close for speed reason.
        * User could stop train when check error, or continue train without
          this train sample.
      * use deque instead of vector in generators pool, make erase
        generator faster.
      * Add chinese/english documentation
      * Make should shuffle = false in unittest
      * Add python files to depends.
      90b9cba7
  18. 19 9月, 2016 2 次提交
  19. 14 9月, 2016 1 次提交
  20. 13 9月, 2016 3 次提交
  21. 12 9月, 2016 1 次提交
  22. 10 9月, 2016 1 次提交
  23. 09 9月, 2016 5 次提交
  24. 06 9月, 2016 2 次提交
  25. 29 8月, 2016 1 次提交