Skip to content
This project is mirrored from https://review.mlplatform.org/tosa/specification. Pull mirroring updated .
  1. Jun 01, 2023
  2. May 25, 2023
  3. May 24, 2023
    • Kevin Petit's avatar
      Formalise the description of operator argument types · 5333c25b
      Kevin Petit authored
      
      
      - Standardise the terminology for operator arguments. Argument,
        Operand, and Parameter were used interchangeably.
      - Introduce a templatized tensor_t<> type for tensor arguments.
        Scalars are represented by rank-0 tensors.
      - Types can be checked with the XSD schema.
      
      Signed-off-by: default avatarKevin Petit <kevin.petit@arm.com>
      Change-Id: Ic57b9387950824e994c5e7f9ec1489c29159b974
      5333c25b
  4. May 23, 2023
  5. May 22, 2023
  6. May 19, 2023
  7. May 18, 2023
  8. May 17, 2023
  9. May 15, 2023
    • Eric Kunze's avatar
      Add TOSA rank requirements to TOSA XML · 42085e36
      Eric Kunze authored
      
      
      Adds new optional element to argument 'rank'
      - Must supply minimum and maximum rank
         - Integer values or the level based "MAX_RANK"
         - trailing modifiers allowed for "MAX_RANK"
      - Displays in a new column in the document
      - Document generation validates rank against specified shape
      
      Change-Id: I507dc51bfe012d3230af43103c6c423a6f1e92b5
      Signed-off-by: Eric Kunze's avatarEric Kunze <eric.kunze@arm.com>
      42085e36
  10. May 09, 2023
    • Dominic Symes's avatar
      Main inference compliance testing updates · 5b936a3c
      Dominic Symes authored
      
      
      - Add additional main inference compliance test
      data sets for dot product testing in Appendix A.
      
      - Express dot product test criteria in terms of
      tensors rather than individual dot products.
      
      - Add per-operation details on test set generation
      in Appendix A.
      
      - Clarify compliance vs conformance wording.
      
      - Clarify that the comment in the table of section 1.8.2
      on integer saturation applies to the CAST from
      floating point to integer operation.
      
      Change-Id: I1b4c4493b02ed7c8a6eb547656c91ca67d4b0e86
      Signed-off-by: default avatarDominic Symes <dominic.symes@arm.com>
      5b936a3c
  11. May 02, 2023
  12. Apr 28, 2023
    • Kevin Petit's avatar
      Fix resize_t enum name · 53e76592
      Kevin Petit authored
      
      
      resize_t is already used in the type support table for RESIZE, rename
      the mode enum to resize_mode_t.
      
      Also add a python function to get an enum object by name to the spec
      object.
      
      Change-Id: Id12c10929beda469c6e3714518d6e26f18ddd440
      Signed-off-by: default avatarKevin Petit <kevin.petit@arm.com>
      53e76592
  13. Apr 27, 2023
  14. Apr 26, 2023
  15. Apr 18, 2023
  16. Apr 14, 2023
  17. Apr 06, 2023
  18. Mar 21, 2023
  19. Mar 20, 2023
  20. Mar 17, 2023
  21. Mar 16, 2023
  22. Mar 08, 2023
  23. Mar 07, 2023
  24. Feb 24, 2023
    • Dominic Symes's avatar
      Main inference conformance additions · 8f7688d7
      Dominic Symes authored and Eric Kunze's avatar Eric Kunze committed
      
      
      Add a check on the error bias for dot
      product calculations and slightly increase the
      error variance bound for dot products.
      
      Change-Id: I518704d1fb1b109695b0f90f7ca6733850e6a8c7
      Signed-off-by: default avatarDominic Symes <dominic.symes@arm.com>
      8f7688d7
    • Luke Hutton's avatar
      Fix rank limits for various data layout operations · 580fdd13
      Luke Hutton authored
      
      
      This commit removes the maximum rank 4 restriction
      from REVERSE, SLICE and TILE operations so that the
      new maximum is now MAX_RANK.
      
      In doing so some rank inconsistencies were also
      cleaned up, including:
      * Adding a minimum rank to the PAD op description
      * Adding level limit checks to SLICE and TILE ops
      * Adding checks for rank(shape1) == rank(shape)
        for TILE and TRANSPOSE ops
      * Add tensor_size(shape1) == tensor_size(shape)
        check for TRANSPOSE
      * Replace tensor name with shape name in pesudo
        code where necessary
      
      Change-Id: I94bb8ee0b6b720b2ba5a884ce9300dca6281245a
      Signed-off-by: Luke Hutton's avatarLuke Hutton <luke.hutton@arm.com>
      580fdd13
  25. Feb 06, 2023
    • Eric Kunze's avatar
      Add accumulator size attribute to AVG_POOL2D · 87bf309c
      Eric Kunze authored
      The accumulator size cannot be determined from the current signature,
      so a new attribute is needed to describe the possible accumulator
      sizes.
      
      Change-Id: I6de81b9ef2bb62a1ca726866f6561c6ae44904aa
      87bf309c
  26. Feb 01, 2023
  27. Jan 23, 2023
  28. Jan 17, 2023
  29. Dec 09, 2022
  30. Dec 05, 2022
  31. Nov 30, 2022
  32. Nov 11, 2022
  33. Oct 19, 2022
Loading