index
:
reference_model.git
main
v0.23
v0.30
v0.40
v0.50
v0.60
v0.70
v0.80
v0.90
v1.0
[no description]
about
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
reference_model
Age
Commit message (
Expand
)
Author
2022-02-01
Rework divisor calculation for AVGPOOL2D
Eric Kunze
2022-01-20
Fix for LOGICAL_LEFT/RIGHT_SHIFT shift values
Jeremy Johnson
2021-11-11
Fix broadcast bug
Kevin Cheng
2021-11-11
More ERROR_IF to check attribute for convolution ops
Kevin Cheng
2021-11-09
Check valid broadcastable shape for binary and ternary ops
Kevin Cheng
2021-11-05
Cleanup CMakeLists.txt
Kevin Cheng
2021-11-04
Update reference model version to 0.24.0 draft
Eric Kunze
2021-11-03
Remove draft tag for 0.23 release
v0.23.0
v0.23
Eric Kunze
2021-11-02
more ERROR_IF fixes
Kevin Cheng
2021-11-01
Fix for tensor_ops.cc
Kevin Cheng
2021-10-28
Changes for 0.23.0 release
Kevin Cheng
2021-10-26
Add version to reference model
Kevin Cheng
2021-10-21
Fix typo in Concat ERROR_IF
Kevin Cheng
2021-10-21
Fix pool2d generation and check to match specification
Jeremy Johnson
2021-10-18
Add ERROR_IF to control flow ops.
Kevin Cheng
2021-10-18
More ERROR_IF supports
Kevin Cheng
2021-10-13
Catch ERROR_IF on AVG_POOL2D and MAX_POOL2D
Kevin Cheng
2021-10-06
Fix reduction ERROR_IF cases
Kevin Cheng
2021-10-05
Couple of reference model fixes
Kevin Cheng
2021-09-30
Fixes to pass NEGATE op test.
Kevin Cheng
2021-09-30
Add SUBGRAPH_ERROR_IF() to catch graph-level error.
Kevin Cheng
2021-09-28
Remove ReluN op.
Kevin Cheng
2021-09-28
Removing rank 0 broadcast in binary op.
Kevin Cheng
2021-09-27
Op that violates rank requirement now runs to the end and return ERROR instea...
Kevin Cheng
2021-09-23
Add stride larger than dimension error_if check for resize
Matthew Haddon
2021-09-20
Add maximum dimension check ERROR_IF in RESIZE.
Kevin Cheng
2021-09-16
Implement Conv3D kernel.
Kevin Cheng
2021-09-15
Rename attribute: Pool2d, Conv2d, TransposeConv2d -> Pool, Conv, TransposeConv
Kevin Cheng
2021-09-14
Add saturation REQUIREs for ADD,SUB,MUL,INTDIV
Jeremy Johnson
2021-08-31
Rename DIV operator to INTDIV
Matthew Haddon
2021-08-20
Replace node level check ASSERT_MSG_NODE()/FATAL_ERROR_NODE() with REQUIRE() ...
Kevin Cheng
2021-08-12
Support int4 weights read. Added conv2d int8xint4 in test generation.
Kevin Cheng
2021-07-01
Adding option -Coutput_dir= to control output location.
Kevin Cheng
2021-06-24
Update to use new serialization_lib API.
Kevin Cheng
2021-06-09
adding batch dimension to MatMul.
Kevin Cheng
2021-06-04
Rewrite model frontend to be json-driven.
Kevin Cheng
2021-05-24
Support 8-bit TABLE op.
Kevin Cheng
2021-05-12
Update to v0.22.0
Kevin Cheng
2021-04-30
Fix OpMul for 32-bit x 32-bit inputs
Kevin Cheng
2021-04-30
Support 16-bit Rescale
Kevin Cheng
2021-04-30
Concat takes variadic inputs
Kevin Cheng
2021-04-28
Support mixed-precision input tensors for TOSA unit test.
Kevin Cheng
2021-04-27
Replace serialization/ and verif/ with MLPlatform's serialization_lib submodule
Kevin Cheng
2021-04-23
Fix bug that NegateOp doesn't clip output value
Kevin Cheng
2021-02-22
Fix operator initialization bug when rank/dtype fails to initialize with Usag...
Kevin Cheng
2021-02-01
Remove AINT8
Kevin Cheng
2020-11-24
Update reference model/serialization library to 0.21.0 with unit tests added/...
Kevin Cheng
2020-11-12
Implement and add unit tests for MUL and ARITHMETIC_RIGHT_SHIFT
Kevin Cheng
2020-10-19
Update apply_scale_32()
Kevin Cheng
2020-10-14
Initial checkin of TOSA reference_model and tests
Eric Kunze