Age | Commit message (Collapse) | Author |
|
Add RankMismatch and DimensionMismatch support for SELECT
Update RankMismatch ops to also support DimensionMismatch
Update POW op to have proper broadcast testing
A few other broadcastable ops missing Rank/Dimension testing
Change-Id: I6566f45a7a0db4f9f008456ea7a8e23d4192f4f9
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
* Transpose WrongRank tests now use ranks 7, 8
* Concat ERROR_IF checks now test for inaccurate summation
of output shape tensor dimension
Change-Id: If32f43a4dbd872d0ef7625fa3d4969c863a11b8c
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Signed-off-by: Les Bell <les.bell@arm.com>
Change-Id: Ie6c8c8653874f9eed6007a54a3ad526601a4a669
|
|
* Operators implemented: sigmoid, tanh, arthmetic_right_shift,
mul, table, select, equal, greater, greater_equal, concat, reverse,
tile, scatter, gather, case
* Note that over the course of implementation some specific ERROR_IF
checks have been added for some of the above operators
Change-Id: I80595e6eb9a3e5efd1cc6fd7aa28bbc2dd614980
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Signed-off-by: Les Bell <les.bell@arm.com>
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
Allow random shift values rather than just 11
Check for values that cause unpredictable behaviour
Change-Id: I508c9d0982808ac03b8400d47ad9445d6afb1b79
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
- update serialization_lib hash
- PAD:
1. make padding as an attribute instead of tensor.
2. add pad_const_int (for non-float type) / pad_const_fp (for float type)
- TRANSPOSE: make perm as an attribute instead of tensor
- TABLE: make table as attribute instead of tensor
- update examples/ tests
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Iddc446db4b356ba2f36ea4a79b7220b9cfc2aa4e
|
|
Change-Id: Idb36b1f1c0d78ec101c168865a9c8d03221b4c84
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
Change output size error_if check in ref model to match specification.
Remove size check from test generation as output shape is always
correctly calculated.
Change-Id: I5be64f31e6448b47e80fc0a4af11bb312f366a26
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I659337aadfd0498bf88a95737f69c51efec797cc
|
|
Change-Id: I75f2a4ab6790dcbdfaec064f42f601d8f44da70b
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
* Negative tests for rescale op added
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I70aead1c6a67f159c7b7c9a05f7d5f0b92521584
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Idd12aab4d52365f3137a403d362790abc16e66d0
|
|
Change-Id: Id1c2d2a41bdfff6b3fe45f79bf99abd0b5beb704
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
* Negative tests for ERROR_IFs given in spec
* Constrict dimension size of latter ranks if
rank is larger than 4
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: Iffea1874e876dba83c8a7c63049283bf7b3ba74b
|
|
* Negative testing support added for the following operators:
reduce_all, reduce_any, reduce_max, reduce_min, reduce_product, reduce_sum
Change-Id: Ife4f0307b681510e8910a23a39efc127c82ae120
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
Change-Id: Ic2aebe40b5cce61d4576a64f4f48ff87b36475c2
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
* Added negative testing support for the following operators:
abs, bitwise_not, ceil, clz, exp, floor, log, logical_not, negate, reciprocal, rsqrt
Change-Id: Icc6f146c6407502520330678420951749ba2a9ef
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
* Negative testing support for the following operators:
ADD, BITWISE_AND, BITWISE_OR, BITWISE_XOR, INTDIV, LOGICAL_AND,
LOGICAL_LEFT_SHIFT, LOGICAL_RIGHT_SHIFT, LOGICAL_OR, LOGICAL_XOR,
MAXIMUM, MINIMUM, POW, SUB
Change-Id: I2271f00b0b619604e864e36e4a4f987f1b2a37d4
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
Change-Id: I7c670c5f9b97a18a6f586b16f31bc9fc301f6bc3
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
* Positive and negative tests are now produced entirely in isolation,
if 'both' test types are chosen all positive tests are generated,
then all the negative tests are generated, without combining the two
* Moved tensor generation out of serialize test and into it's own function
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: Id8e9023d2c2966a0b14ae83d2d848a66cb125fcb
|
|
Change-Id: I2a0aa63a4256629d12166638812dc5b854db2ddf
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
* Added WrongInputType and WrongOutputType checks
* Delete and Add extra inputs/outputs when doing op graph checks
* Add incorrect rank checks
* Entire op dictionary is now passed to build functions
* Error if validation now done in separate function
* Duplicate tests are ignored
Change-Id: Ie1117bdbc9bdeb42059123792ce1df4cc56db54e
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Iea91076f3e0673ce5adfcca247d968416023fa58
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Ic0c4bd1a55c6782aeacb77163a3a093edb86d437
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Iaac727159a84de1f83de549c3a22704096f46bf9
|
|
* TosaErrorValidator implemented to produce and test for
ERROR_IF conditions
* RESIZE specific ERROR_IF test support added
* Set rank and type parameters before test generation loop to
avoid multiple checks for valid parameters
* Increase output dimensions if IFM/OFM ratio smaller than 1/16
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I430e13383d99c2e25354f53d3703fb9be973f6d4
|
|
* larger values for stride/padding/dilation/kernel
* sparse test generation, as there are too many variations
Signed-off-by: Les Bell <les.bell@arm.com>
Change-Id: If13ea17024d81262ab892e3111cbf5833e77a8c5
|
|
- Catch and classify as INTERNAL_ERROR unix signals thrown by
reference model
- Made -t a required argument
Change-Id: I2be1ed6f5c1e862d6a3aee43c6b1e8575ab50d1f
Signed-off-by: Jared Smolens <jared.smolens@arm.com>
|
|
- Update the serialization code to work with both
flatbuffers 1.12.0 and 2.0.0 APIs
Change-Id: If41da785a2ab756de94b523d04358489ef01d603
Signed-off-by: Jared Smolens <jared.smolens@arm.com>
|
|
* plus extra debug info for some exceptions
Change-Id: I667ee9b8a1cba8c6bd47c31231f1805eba680ba5
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Ic16e918b1a2423ad563684e29ce70d9efdbf9c02
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: I466dd1dcf5230e8e07df202ba88515e775e04a1e
|
|
Change-Id: I4f05b256f4f439f72e5ee8bce60e4e92b6aaa6e7
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|
|
- add output tensor to main block
- fix bug where tensor cannot serialize list of const data
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: I40b96d634642e24e6fae6fa3ad6846f5911005f7
|
|
* Implemented InvalidValidator to remove existing invalid tests.
* Removed invalid tests for resize, rescale, conv2d, depthwise_conv2d,
transpose_conv2d, avg_pool2d, and max_pool2d (note default avg/max_pool
never produced negative tests, but theoretically could).
* Changed behaviour of computerMultiplierAndShift to produce the allowed
range of shift values.
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I5e7b11030deb5322e2ca08fd4f4467fb02b7740d
|
|
* The option --test-type allows the user to select 'positive', 'negative',
or 'both' types of tests produced by the test generator.
* Reset RNG when looping through negative test generation (generation not
implemented)
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I1bfcb3170e7380be0f98b36b3d4abc4779a05abe
|
|
* max batch size ignored if target shape is set
* W size reduced if large input tensor used
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I13472ab768fa93a1d0b9e28964f56ec4a06dbdfd
|
|
* In line with the TOSA spec the DIV operator has been renamed INTDIV
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: I1dc6e88220ef26b24487675600b6cd1e5bb8b0f7
|
|
* Sets multiple of input dimension to one if the dimension of the
input tensor is large to reduce ouput tensor size.
Change-Id: I2ebcecf438282de032a33e0cf4b3847cbf94440d
|
|
- Updated example tests, added missing npy files
- Updated README file with fields in desc.json
Change-Id: I5651ed5260866a8381adf620fb6dd6405cb2833b
|
|
or ERROR_IF()
- Adding return code enum class: {VALID, UNPREDICTABLE, ERROR}
- Runtime errors (e.g. memory allocation failure) will abort immediately, or will return one of the three return codes
Part of the codes are re-written to pass REQUIRE() to the top-level (e.g. apply_scale_32/16())
- Update setExpectedFailure() to setExpectedReturnCode() on test generation script
- Update test regression script to interface with reference model change
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: Ia063c936bcb2a54d6e379a5bb6801aa72d1186f1
|
|
* Currently when a target shape is specified, the W value of the input
tensor is always equal to N, this is not the case when no target shape
is defined.
* A random value for W is generated every time.
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I8f8ecb32308cef4a1ece1871f76ebbd5f0cf881f
|
|
* Concat tests now contain between 2 and 5 input tensors concatenated
together
* Both input and const tensors are used as inputs to the operator
* Option to add in const tensor inputs (this is slow), defaults to
original behaviour
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I2a0cc622d31aceab8d24521668d0aae040ba73b1
|
|
Signed-off-by: Kevin Cheng <kevin.cheng@arm.com>
Change-Id: I61620f160c7dad6aac5fcc3da0a6e97f3bae5b40
|
|
* encode pad in the test name with leading zeros
Change-Id: I7b1cc04014d390e707de099c5f78f93f5f344d69
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
* RESCALE can now produce tests with UINT8
as the input/output datatype.
Signed-off-by: Matthew Haddon <matthew.haddon@arm.com>
Change-Id: I0a5d7b3c6dd7c2501d14e5d558b1f18e5e967fa9
|
|
Change-Id: Ib299b78abe0e5c2ef0aeff9d853e583078768e1c
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
Signed-off-by: Les Bell <les.bell@arm.com>
Change-Id: I9daf8fd691af74500819718519caac30c966a1f2
|
|
* for int8 and uint8 only
* not for int16
* pass all data types to the quantizaion generator functions as some
operators (all convolutions, fully_connected) may use different types
for the input and weight tensors
* also allows input type filtering for convolution operators
Change-Id: Iea3d00f03807a8db35b40d4e8988929ec6549b44
Signed-off-by: Les Bell <les.bell@arm.com>
|
|
Change range of INT4 to -8 to 7 and INT8 to -128 to 127.
Change clamp attributes to be in range of signed input type.
Change-Id: Iccee1bf29b12da6f7622922f89d59cc0730460c2
Signed-off-by: Jeremy Johnson <jeremy.johnson@arm.com>
|