1. 11 Jun, 2019 1 commit
  2. 09 May, 2019 1 commit
    • Robert Kimball's avatar
      Store Constant arrays where data is constant as a single value (#2880) · 7c337e5d
      Robert Kimball authored
      * wip
      
      * Add support for storing constant array of constant values as a single values that is automatically broadcast on deserialize
      
      * revert some changes to serializer.cpp
      
      * fix all_close_f to support nan and inf to allow for unit test
      
      * update unit tests to pass for all_close_f update
      
      * fix bug with i64
      
      * address compile issues?
      
      * change function name to be more accurate
      
      * fix compiler error
      7c337e5d
  3. 25 Mar, 2019 1 commit
    • gcwenger's avatar
      Added "min_signal" paramater to float testing (#2653) · abd1c70d
      gcwenger authored
      min_signal is intended to optionally skip checking float
      distances when numbers are close enough to 0.
      Only skips when both numbers are < min_signal.
      Intention is to allow tighter float testing in certain
      cases where most values are not near 0, but values near
      zero are differing by more bits than values farther from 0.
      Should be used with caution in limited cases.
      abd1c70d
  4. 11 Mar, 2019 1 commit
  5. 07 Jan, 2019 1 commit
    • gcwenger's avatar
      Simplified all_close_f interface and tightened default criteria (#2285) · 0eaa960c
      gcwenger authored
      * Simplified & tightened all_close_f parameters
      
      Removed specification of mantissa bits for all_close_f in favor
      of just specifying tolerance bits. Tightened up all_close_f default.
      Fixed LRN unit test which had insufficient result precision to pass
      tighter all_close_f tolerance.
      
      * Addressed PR comments.
      
      Reworked mantissa bit and tolerance constants.
      Clarified and improved graph comparison tolerance calculation flexibility.
      Clarified unit test tolerance testing.
      0eaa960c
  6. 03 Jan, 2019 1 commit
  7. 19 Dec, 2018 1 commit
  8. 13 Dec, 2018 1 commit
    • gcwenger's avatar
      Addressed some all_close_f loose ends (#2200) · 656b2f72
      gcwenger authored
      * Simplified all_close_f tolerance determination. Made all_close_f less verbose. Transitioned graph comparison to use double comparison when appropriate.
      
      * Reworked all_close_f AssertionResult returning code to include message in AssertionResult on success also.
      
      * Added CMake option NGRAPH_GTEST_INFO, which controls whether or not any gtest info cout is done. Quiets all_close_f tests by default.
      
      * Moved NGRAPH_GTEST_INFO from build option to environment variable.
      656b2f72
  9. 08 Dec, 2018 1 commit
  10. 07 Dec, 2018 1 commit
    • gcwenger's avatar
      Support for all_close_f w/ doubles (#2184) · 125f7242
      gcwenger authored
      * Double support for all_close_f
      
      * all_close_f uses fixed number of mantissa bits now. Simplified testing code.
      
      * Initialize test data members in constructor to values which will cause test failure. Setup then sets them correctly.
      
      * Reduce info printed out during all_close_f unit tests.
      125f7242
  11. 06 Dec, 2018 1 commit
    • Fenglei's avatar
      nvgpu cuda reduce with stable sum (#2076) · 606f3f93
      Fenglei authored
      * add some helper function
      
      * update with new helper function
      
      * update reduce to nd with new helper function
      
      * update float sum to stable sum
      
      * fix bug
      
      * update all reduce to stable sum for float
      
      * fix bug and pass the sum stable test
      
      * remove debug info
      
      * style
      
      * update with shape
      
      * fix bug
      
      * add host parameters to cuda_emitter
      
      * clang format
      
      * fix bugs
      
      * add element::type support
      
      * format
      
      * add a cached value with datatype name
      
      * add init_reduce_value
      
      * unroll loop
      
      * optimization
      
      * remove the need for init_value
      
      * add memset kernel
      
      * add memcpy
      
      * working version
      
      * remove debug info
      
      * add comments, clean up code.
      
      * change in_idx to input_idx
      
      * fix bug
      
      * change args name for memset in emitter
      
      * pass element::Type instead of string
      
      * the op::reduce come with init value, add support
      
      * resolve codacy-bot comment
      
      * fix bug
      
      * resove codacy-bot comment
      
      * remove unused comments, resolve comments
      
      * cuda reduce for max, min, mul, reduce op init value, format
      
      * use type::info
      
      * use type info for numeric_limits
      
      * remove code from gpu_host_parameters
      
      * header
      
      * remvoe outdated comments
      
      * add helper to check if stable sum is needed
      
      * add stable sum test for double
      
      * remove extra line
      
      * consolidate helper functions
      
      * no need list now.
      
      * remove extra ;
      
      * clang format
      
      * style
      
      * add skip test for cpu and intelGPU side
      
      * add line between groups of headers
      
      * add two simple stable sum test for float and double
      
      * skip test for intelGPU
      606f3f93
  12. 01 Dec, 2018 1 commit
  13. 11 Nov, 2018 1 commit
    • Fenglei's avatar
      add isfinite check for all_close (#2028) · 702d465a
      Fenglei authored
      * add isfinite check
      
      * style
      
      * output 5 diff and total diff
      
      * output limit of diff for all_close_f
      
      * dix bug
      
      * disable tests
      
      * remove failing unit test that does not make sense.
      702d465a
  14. 10 Nov, 2018 1 commit
    • gcwenger's avatar
      Heterogenous serialized graph testing across backends (#2020) · 40bcfdf7
      gcwenger authored
      * Heterogenous sub-graph comparison testing
      
      * Print index for float differences
      
      * Disabled compare_backends_with_graphs on most backends for now. Moved to new file. Added testing of unsigned values.
      
      * Fixed element::boolean range. Added missing include.
      
      * Switched use of shared_ptr as parm to raw *. Moved to using namespace std in cpp. Fixed comment marker in unit_test.manifest files. Switched some EXPECT_EQ TO ASSERT_EQ. Fixed parameterized test disabling.
      
      * Frozen naming -> serialized. Removed extraneous comments.
      
      * Graph comparison unit test relies on CPU for reference, so only build when CPU is built.
      
      * Reworked per backend disabling of compare_backends_with_graphs
      40bcfdf7
  15. 29 Sep, 2018 1 commit
  16. 14 Sep, 2018 1 commit
    • Scott Cyphers's avatar
      Cyphers/layout (#1602) · 2f79f707
      Scott Cyphers authored
      * Remove "view"
      Simplify layout
      
      * Fix merge error
      
      * fix build error
      
      * PR1602. IntelGPU backend. Compilation fixed.
      2f79f707
  17. 29 Aug, 2018 1 commit
  18. 02 Jun, 2018 1 commit