Age | Commit message (Collapse) | Author |
|
Added Pseudo Random number generator in generate library.
Enabled MAX_POOL2D, PAD FP32 tests to use new generator and compliance.
Fixed verify library exact mode to expect reference data as FP64.
Simplified tosa_verif_build_tests internal interfaces for new tests.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: Icc0ffa924cf38107c3a212efd452c47a650c9d98
|
|
Improved libraries validation to catch unknown values.
Improved verify output to match generate library.
Refactored generate tests to use less code duplication.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: I9c38745fbc8e70f46c19ddae6c62ee248d33b5f1
|
|
CONST data should be either data in the flatbuffer file or loaded
in via an input file. This check catches if neither of these have
been done, instead of marking uninitialized data as valid.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: I97dc2254f0b58c05c39cc0281a331a394c2a4b3c
|
|
Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: I414899d0f504af00da185e0fc4119f3bde2bae3a
|
|
- Nested looping instead of using Eigen extract_image_patches to avoid OOM errors
Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Id4d78d5b5dd04a00134f29b1d29f814195515b1f
|
|
Add support for tensors of booleans as input arguments to operators.
Signed-off-by: Jack Frankland <jack.frankland@arm.com>
Change-Id: I35bb5768cef11348c27ffbe5869c32c144485958
|
|
Added support for using generate library in tosa_verif_build_tests
and tosa_verif_run_tests tosa tool scripts.
Reduced scope of compliance test creation and verification to
the supported type of FP32.
Fix missing virtual destructor warning in generate_dot_product.h and
add config file for generate library.
Simple pytests included to check python interface.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: I6cdad9b00660d6ddc8bd07fdea813937fb48626a
|
|
Fixed PrimitiveGenerator starting point and added test sets 1-5.
Fixed verify_test reduce_product missing data_type.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: Iaf080ce5c1adb5819f70d1a285d04baa36016092
|
|
Add support for dot-product MatMul - test set 0
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: Ifd15b42570014b634f59c94a1fd1cd56bac79ea4
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
* Add verifiers to validate the result of a reduce produce operation.
* Add test cases for the new validator.
Change-Id: I666d1a67f498e7893e0f224bc5408a4134f2ef6c
Signed-off-by: Jack Frankland <jack.frankland@arm.com>
|
|
Updated to conformance generator to not generate tests with results for
compliance tests.
Updated test runner to run compliance mode version (precise & abs mode)
of reference model to create test results to use against SUT results.
Updated reference model to enable abs_mode on correct desc.json flags.
Updated test checker to support compliance checking using verifier lib.
Seperated color printing from test checker.
Change-Id: I7e2fbfc6883916caa5d94d4ece122c48bf45f530
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
Make compliance meta-data data-type required for all.
Add data-type checking for all verifier modes.
Add initial enum support for new ROUND compliance mode.
Improve print out information from library.
Use numpy ctypes.data_as to get f16 support compared to ctypes_lib.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: Ie983ba4ea958a88556f30c09b3ebc19cd9ec96b7
|
|
Add a verifier to check two results are correct within a certain ULP
tolerance for IEEE-754 32-bit floating point values.
Add a test to check the ULP verifier is correct.
Signed-off-by: Jack Frankland <jack.frankland@arm.com>
Change-Id: Iaf43069f300999479d998e7837746b247ca5177e
|
|
Add a verifier to check that tensor results match exactly.
Add a unit test to check the behavior of this new verifier.
Change-Id: I9b80a6d57640fec67c6be80a97b3af484aeb935e
Signed-off-by: Jack Frankland <jack.frankland@arm.com>
|
|
Signed-off-by: Dmitrii Agibov <dmitrii.agibov@arm.com>
Change-Id: I4cd818af0db5e6e8a96641246cd3263ba4878f56
|
|
MAX_SCALE for the EIGHTK level is 256 now.
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I6f3a3623d18ff238888d994c2a4129091f5a92fe
|
|
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: I03cb878736ccd7e1f5e1f780d7171949a19a9de2
|
|
Include 0.90 draft serialization library
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I290360bfd1c147507e092868b5546c9be08682c2
|
|
- Updates TOSA specification to the latest version
- Updates generate_api.py to generate the operator API correctly
for ops with additional tensor inputs.
- Removes default arguments for func_debug and func_config to make
the API C compliant again.
- Updates model_runner_tests.cpp for operators that have changed.
- Adds a unit test for the Tile operator to check that generated code
for additional tensor inputs works correctly.
Signed-off-by: Grant Watson <grant.watson@arm.com>
Change-Id: I1e26065c6ed333b2ca4b3da39972d30f896fa6e5
|
|
Build steps to create tosa_reference_verify_lib.so
New python class to handle loading and calling verifier with pytests.
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: Id42c71c398ed7364742b4a039977412ef3081c4a
|
|
Align with the change in the spec.
Define idiv_floor to give equivalent behavior to the floating-point
floor function for image coordinate calculation.
Change-Id: Ice06d354d58b1bb0dfedab55c9cc9eac1598b50c
Signed-off-by: TatWai Chong <tatwai.chong@arm.com>
|
|
input_unsigned and output_unsigned were added to the specification.
Older TOSA files with uint data types are still supported.
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I125886ffc92975d99971e56e2075dd5d96bdbdc4
|
|
Change verifier API to consume verification configuration in a JSON
format and enable appropriate validation to be performed within the
verifier code in the reference model.
Also update to latest spec changes for main compliance but not yet
including bias support.
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: I0ceaa1714dd041b00b5b29cd937c8f05e701bc4c
|
|
Serialization_lib change: https://eu-gerrit-1.euhpc.arm.com/c/mltech/mirrors/tosa/serialization_lib/+/542261
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: I59752a3ba7981c356634b911160cedf30fb2599e
|
|
MacOS does not have the `sys/prctl.h` header. This patch conditionally
excludes `prctl.h` on MacOS and usage of its functionality in the same
way we currently do for windows.
Signed-off-by: Jack Frankland <jack.frankland@arm.com>
Change-Id: Ic56ec358552126b3a10827d9c52db388a8acc214
|
|
Updates the generate_api.py script and associated
templates to allow func_config and debug_config
to be passed when running individual operators
on the API.
This will allow us, for example, to set precise_mode
and abs_mode when running individual operators.
Signed-off-by: Grant Watson <grant.watson@arm.com>
Change-Id: Ia3e7ffc146f876daa307558433177c68285843b7
|
|
remove unnecessary change of tensor shape array for SHAPE data type
Signed-off-by: Tai Ly <tai.ly@arm.com>
Change-Id: I8e6ca98448188a1c30f9d29c22b3194d327b17ac
|
|
Signed-off-by: Won Jeon <won.jeon@arm.com>
Change-Id: Id5820033bb3cc98b5e284237a91b22ffd54a80e3
|
|
Signed-off-by: Won Jeon <won.jeon@arm.com>
Change-Id: Iea11ee5d3d98773e9c5e9b827593c05afb41ce3b
|
|
add support for bias broadcasting for operators:
- conv2d
- conv3d
- depthwise_conv2d
- transpose_conv2d
- fully_connected
could not add framework test for this because tf.nn.bias_add requires
bias size to match channel dimension.
manually tested reference model evaluation on tosa mlir with bias size
of 1
Signed-off-by: Tai Ly <tai.ly@arm.com>
Change-Id: I70d29d231a63fc03b10e3006cbd6b16b53cca1f2
|
|
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I22939e75a08349816a9c103eb7312edcec53855e
|
|
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I8645382983c257e5102982d487283560088e2d2a
|
|
Signed-off-by: Jiacheng Liang <jiacheng.liang@arm.com>
Change-Id: I473adc1525319b5574ee0e36d10a530277d9215d
|
|
- The previous ref_model was allocating the memory for all tensors in
the graph upfront which is unnecessary and wasteful.
- This patch changes to only allocate initial input tensors on the entry
point using the allocateInputTensor() function.
- The output tensors are ensured to have been allocated before executing
a node. The output tenosrs are the inputs for the next node.
- When a node's evaluation is finished, its input tensors will be freed
if they will no longer be used by anyone else.
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: Ibb3e8c9e6344f6cd9eb20532a03b2097b93247f9
|
|
- Also added run clang-format to pre-commit runs
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: I4e59ac0afbaa30dce0773aa63d92a1a3b119e2f3
|
|
Also move a large amount of redundant code to a macro
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: Ib1dcb2ce15e56076e444a4739c4d1292389c4a78
|
|
Signed-off-by: Jiacheng Liang <jiacheng.liang@arm.com>
Change-Id: Ie31041cbcd01f6290be06eabfe5c619123ae989f
|
|
Would cause unresolved symbols to appear when building in
release mode.
There are a couple of minor compiler warning fixes as well.
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I0f7e9a8771442a6e3c848edfe034ef534d0d8ca7
|
|
Signed-off-by: Won Jeon <won.jeon@arm.com>
Change-Id: Ib42b867287b83a183a0d0fb1f1eb29974f58fae4
|
|
Signed-off-by: Grant Watson <grant.watson@arm.com>
Change-Id: I1296f968baca335ea88691bc973e2d01b2aa2c5b
|
|
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I45287af6b33cf07383250dc962a97728e65697a7
|
|
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: I7460ad9eed3ed5c7cec6e855a0303753ed28eb1c
|
|
Adds utility functions that enable compliance verification of TOSA MI
operators; as per section 1.8.2 in the TOSA specification.
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I0cced0ff8875ac8d78b1943211438713d1c51b88
|
|
- Allows axis attributes to be treated differently to other arguments in attribute.def
Signed-off-by: Grant Watson <grant.watson@arm.com>
Change-Id: I1be2595c24bf22e5391a2911a5283391d310df37
|
|
Comparison operators produce boolean outputs, which need to be written into client data
Allow subgraph traverser to use main block to look for tensors when serialization handler is missing
Signed-off-by: Jiacheng Liang <jiacheng.liang@arm.com>
Change-Id: I6f9af470185541fa6466b3f7786c48f1555fa6f6
|
|
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I175ef0acc6a6633c3e351b935a04db16de1bfbd5
|
|
Update serialization library submodule
Signed-off-by: Eric Kunze <eric.kunze@arm.com>
Change-Id: I1d6e2700ac55f429de742931627b07c9a4a46f14
|
|
This adds a second run of reference model under precise_mode
when test_desc.json contains a "compliance" dictionary which
contains a "mode" entry with value "dot product".
In this second run, abs_mode will be set to true, which causes:
1. evaluation will take absolute values of inputs for these operators:
conv2d, conv3d, depthwise_conv2d, fully_connected,
matmul, transpose_conv2d, fft2d, rfft2d
reduce_sum, avg_pool2d
2. output files will have prefix "bounds_" prepended to them
Signed-off-by: Tai Ly <tai.ly@arm.com>
Change-Id: I7070ecc7ead2d2ea3375c44663d653c6772b88e0
|
|
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Change-Id: I0ef7607f4266296a1204c5cccdb5be36f345b5ba
|
|
Signed-off-by: Jerry Ge <jerry.ge@arm.com>
Change-Id: Iad035b31d5e5e83040068e6311501490765bfff7
|