1. 08 11月, 2017 3 次提交
  2. 07 11月, 2017 3 次提交
    • Y
      Add unittest, backward of array read/write op (#5409) · 6cde889b
      Yu Yang 提交于
      * Use stable_sort in lod_rank_table
      
      It is easy to debug and test when use `stable_sort`and the time
      complexity is not changed.
      
      * Add LoDTensorArray
      
      * Stash
      
      * Better debug message for IsInitialized
      
      * Stash
      
      * Better debug message for IsInitialized
      
      * Complete array read/write op unittests
      
      * Add unittest, Gradient of array read/write
      
      * Follow comments
      6cde889b
    • Y
      Update lod_tensor.md (#5383) · 70154597
      Yang Yang(Tony) 提交于
      An important change on lod tensor indexing. A higher level offset will be based on its next level rather than an absolute offset.
      70154597
    • Y
      ReadFromArray/WriteToArray op (#5407) · c9b57dcc
      Yu Yang 提交于
      * Use stable_sort in lod_rank_table
      
      It is easy to debug and test when use `stable_sort`and the time
      complexity is not changed.
      
      * Add LoDTensorArray
      
      * Stash
      
      * Better debug message for IsInitialized
      
      * Stash
      
      * Better debug message for IsInitialized
      
      * Complete array read/write op unittests
      c9b57dcc
  3. 06 11月, 2017 3 次提交
  4. 05 11月, 2017 1 次提交
  5. 04 11月, 2017 2 次提交
    • Q
      Add acc test to image classification (#5336) · 906e2565
      Qiao Longfei 提交于
      * add acc layer
      * memory log level change from 3 to 10
      * use gaussian random to init conv parameters
      * use initializer
      * fix import
      * batch_norm use helper to create persistable var
      * refine code
      * train only 2 batches for test
      * use g_program and g_init_program
      * use XavierInitializer to init fc parameter
      906e2565
    • Y
      Add LoDRankTable (#5349) · 74849158
      Yu Yang 提交于
      * Add LoDRankTable
      
      LoD Rank Table stores the `level` of `lod` which is ordered by sequence
      length in descending order. It is useful when implement dynamic RNN and
      is shared by dynamic RNN memory, dynamic RNN slice input and dynamic
      RNN slice output operators.
      
      * Add InferVarType
      74849158
  6. 03 11月, 2017 1 次提交
  7. 02 11月, 2017 1 次提交
    • Y
      Rewrite StaticRNN with Executor (#5224) · 0a32e74d
      Yu Yang 提交于
      * Init commit
      
      * Make executor use ProgramDescBind
      
      * Change Attribute from BlockDesc to BlockDescBind
      
      * Since we will get the program desc in RNN, just BlockDesc is not
        enough.
      
      * Add DeviceContext to Executor API
      
      * Rewrite RNN
      
      * Pass Python
      
      * AddBiasOp does not care num_flatten_dims
      
      * Stash
      
      * Fix MacOS Compile
      
      * Pass RNN forward
      
      * add python test
      
      * refactor test
      
      * Make compile pass
      
      * add gradopmaker
      
      * First draft done
      
      * Polish code
      
      * add grad op maker and grad infershape
      
      * Polish code
      
      * Fix backward.cc bug
      
      * Fix infershape
      
      * Rename function
      
      * add backward test
      
      * simplify recurrent test
      
      * Update
      
      * Pass unittest
      
      * Add comments & refine test
      
      * Add comments
      
      * refactor test
      
      * Complete Unittest
      
      * fix StepScopes enforce
      
      * Remove unused unittest
      
      * no type error
      
      * Update
      
      * Make RNN Pass unittest
      0a32e74d
  8. 01 11月, 2017 3 次提交
  9. 31 10月, 2017 3 次提交
  10. 30 10月, 2017 2 次提交
  11. 29 10月, 2017 3 次提交
  12. 28 10月, 2017 2 次提交
    • F
      Python API for inference model saving/load (#5020) · 6783dcee
      fengjiayi 提交于
      * Add `dump_to_file()` for ProgrameDescBind in pybind
      
      * Update
      
      * Add utility.py
      
      * typo
      
      * Fix bugs
      
      * Move add_feed/fetch_components to untility.py
      
      * Compelete dump
      
      * Follow comments
      
      * Change output of Prune() from inference to pointer
      
      * Expose Prune() to Python
      
      * Compelete save/load API of inference model
      
      * Fix errors
      
      * Debuging
      
      * Compelete unit tests
      
      * follow comments
      6783dcee
    • Y
      Add debug logs in scope, meta_cache and memory (#5170) · 2a5edec0
      Yu Yang 提交于
      * Add debug logs in scope, meta_cache and memory
      
      * Add missing deps
      2a5edec0
  13. 27 10月, 2017 7 次提交
    • C
      follow comments · cadee843
      chengduoZH 提交于
      cadee843
    • C
      follow comments. · bce4f7d6
      caoying03 提交于
      bce4f7d6
    • Y
      Make InferShape as a field in OpInfo (#5139) · b44f4ccb
      Yu Yang 提交于
      * Op developer can add `InferShape` to any operator
      b44f4ccb
    • Q
      add sparse support for sum op (#5093) · 7f8574c0
      QI JUN 提交于
      * add sparse support for sum op
      
      * typo fix
      
      * fix gpu build error
      
      * fix unittest error
      
      * typo fix
      
      * infer var type and shape in op_test
      
      * follow comments
      
      * fix build error
      
      * bypass some unittests depend on NetOp
      7f8574c0
    • Y
      Gradient check use graph (#5027) · be00b0c4
      Yu Yang 提交于
      * Simplize Gradient Check
      
      * Stash
      
      * Extract apply_backward_pass to backward.py
      
      Rename apply_backward_pass to append_backward_ops
      
      * Use graph API to check gradient
      
      * Fix ci
      
      * Fix CI
      
      * Fix backward for double precision
      
      * Stash
      
      * Fix CI
      
      * Fix ci
      
      * Ignore GRU test
      
      * Ignore xe op
      
      * Fix CI
      
      * Fix softmax with xe gradient
      
      The correct equation should be IG = OG * (d_softmax_with_xe())
      
      * Fix typo
      
      * Fix merge error
      
      * Disable LRN
      be00b0c4
    • D
      "fixed based on comment" · 6cce5268
      Dong Zhihong 提交于
      6cce5268
    • F
      Add functions of restoring ProgramDescBind from ProgramDesc (#5109) · aa379ccb
      fengjiayi 提交于
      * compelete restoring program_bind from program_desc
      
      * Fix bugs
      
      * fix compile errors
      
      * fix errors and add unit tests
      
      * rename some vars
      
      * Follow comments
      aa379ccb
  14. 26 10月, 2017 5 次提交
  15. 25 10月, 2017 1 次提交