summaryrefslogtreecommitdiff
path: root/nn/runtime/test/TestValidateOperations.cpp
AgeCommit message (Collapse)Author
2020-06-10Relax control flow boundary operand dimension constraintSlava Shklyaev
Also updates the NDK spec to mention the constraints and adds some validation tests. Bug: 132458982 Bug: 156918813 Test: NNT_static Change-Id: Ia112e46da065a623a52ac1c402d28dcb963e5580
2020-05-14Fix incorrect (and inverted) comparison.Stephen Hines
This test was incorrectly passing this comparison before because the uninitialized value in `that.channelQuant->scales` was 0xaaaa... instead of not needing to be compared at all when `channelQuant` isn't set. Now we check whether to examine `scales` at all. Bug: http://b/156464649 Bug: http://b/156514991 Test: atest CtsNNAPITestCases:TensorRankConstraint Change-Id: I1f4db99bdd63e5dc1c95f574e6eca3adfffd138d (cherry picked from commit 63deac05d3a1d53430966a49c6979554f128383d)
2020-05-06Improve shape validation in FULLY_CONNECTEDSlava Shklyaev
1. Adds bias dimension count validation. 2. Does more validation in validate(). Bug: 155575142 Bug: 155261461 Test: NNT_static --gtest_filter="*while_fib*" Test: NNT_static --gtest_filter=ValidationTestDimensionProductOverflow2.DynamicOutputShapeOverflow Change-Id: I7b70a29e76fdf99e656ee6e3867cfd97675cfeec
2020-04-15Fix failures in TensorRankConstraint TestsStefano Galarraga
Fix sporadic failures in CtsNNAPITestCases.TensorRankConstraint tests. Improves description of expectation failures when comparing OperandTypeWithExtraParams instances. Test: CtsNNAPITestCases Bug: 154093710 Change-Id: Ia2b8bc914892f2ff7908d247fa26743fa242f783
2020-04-06Add rank check to LOCAL_RESPONSE_NORMALIZATION validationLev Proleev
Also move LOCAL_RESPONSE_NORMALIZATION operation to OperationResolver Bug: 124041202 Bug: 139957496 Test: NNTest_static Change-Id: I4226bdb101d6ee0a744448e61357344db019dd70
2020-04-06Add rank check to FLOOR validationLev Proleev
Also move FLOOR operation to OperationResolver Bug: 124041202 Bug: 139957496 Test: NNTest_static Change-Id: I59070aff36ddfa57085258974168a314176ed27b
2020-04-06Add rank checks to validation functionsLev Proleev
The change adds rank checks to validation of operations that only support tensors of rank 4 or less. This requirement comes from legacy TF Lite code and is likely to be relaxed in the future to be on par with TF Lite. Adding the checks to validation is benefitial for the TF Lite delegate since in case of a validation error NNAPI node will be fully rejected by the delegation but execution error will cause TF Lite to run NNAPI node during every invocation only to receive an error and do the calculation using CPU implementation. Bug: 139957496 Test: NNTest_static Change-Id: I5cc4c48e775826a237d5ac54c3d2078254bd17a2
2020-03-30Add rank validation to CONV_2D.Stefano Galarraga
It also adds mutation testing for rank validation to TestValidateOperations.cpp. Test: TestValidateOperations Bug: 147106551 Change-Id: Ia588772047d6afd0ff51ae434f2816ffcfdeca62
2020-02-05Add a state output for BIDIRECTIONAL_SEQUENCE_RNNLev Proleev
Bug: 138443991 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: Iabda8320facc253c273ea6edb9d0b24b3715c4f9
2020-02-05Add a state output for BIDIRECTIONAL_SEQUENCE_LSTMLev Proleev
Bug: 138443991 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: I4f3292a2426093a6534567812dffb71e126f21f9
2020-02-05Add a state output for UNIDIRECTIONAL_SEQUENCE_LSTMLev Proleev
Bug: 138443991 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: I48b4eeeebf9a317efa2e05ee2caddb168bde5bb8
2020-02-05Add a state output for UNIDIRECTIONAL_SEQUENCE_RNNLev Proleev
Bug: 138443991 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: I87c4735b9ced983142cac502635c1a6a21deed0c
2020-01-22Add FILL and RANK opsLev Proleev
Fix: 148050168 Fix: 148049333 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: I5438cdd7c0cc52e49067529edcee537278c8f7ad
2020-01-22Add HARD_SWISH opLev Proleev
Fix: 147482068 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: I4436d47ec3b2f51ef9c17b0e061c065e92c410a0
2020-01-22Add ELU opLev Proleev
Fix: 147481241 Test: NNTest_static and VtsHalNeuralnetworksV1_3TargetTest Change-Id: Ia6596dc9517a3c17d2b56439df658b1dca96d7dc
2020-01-13Add TENSOR_INT32 support for ABSSlava Shklyaev
Bug: 142318767 Bug: 136735929 Test: NNT_static Change-Id: Iede89032f584c94c7083069280c9afe1abb5df1e
2020-01-13Add TENSOR_INT32 support for DIVSlava Shklyaev
Bug: 142318767 Bug: 136735929 Test: NNT_static Change-Id: I105222e43802bc58562bb904bd9e9b306eb3697c
2020-01-13Add TENSOR_INT32 support for MULSlava Shklyaev
Bug: 142318767 Bug: 136735929 Test: NNT_static Change-Id: I42d1d5874be6bc7ff11ecf33fd15ce9db539d1a3
2020-01-10Merge "Add TENSOR_INT32 support for SUB"Slava Shklyaev
2020-01-10Merge "Add missing OperationValidationTest for signed quantization"TreeHugger Robot
2020-01-09Add TENSOR_INT32 support for SUBSlava Shklyaev
Bug: 142318767 Bug: 136735929 Test: NNT_static Change-Id: I10e7e93d5d58d2f53709ad7f993e10638d80cf56
2020-01-09Implements Quantized LSTM op for R.Viet Dang
Also adds support for TENSOR_QUANT8_ASYMM_SIGNED in Test Generator. Bug: 144841609 Bug: 145916330 Test: NeuralNetworksTest_static Change-Id: I14b0d284b1945833d532cbaa33c66e4d77afd8b7
2020-01-09Add missing OperationValidationTest for signed quantizationLev Proleev
Tests added for: * LOGISTIC * RELU, RELU1, RELU6 * HEATMAP_MAX_KEYPOINT * TANH Bug: 136735770 Test: NNTest_static Change-Id: If67ddee369f48735b664fe63c7dcc991bf4976db
2019-12-16Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for generate proposals ops"Lev Proleev
2019-12-16Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for ROI_ALIGN"Lev Proleev
2019-12-13Add TENSOR_QUANT8_ASYMM_SIGNED support for generate proposals opsPrzemyslaw Szczepaniak
Updated: - GENERATE_PROPOSALS - AXIS_ALIGNED_BBOX_TRANSFORM - BOX_WITH_NMS_LIMIT Fix: 143934558 Fix: 143934723 Fix: 143934469 Test: quantization coupling tests in CTS and VTS Change-Id: I395367bb401c4c596dece90e7080a6c97c7ff3f3
2019-12-12Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for EMBEDDING_LOOKUP"TreeHugger Robot
2019-12-12Add TENSOR_QUANT8_ASYMM_SIGNED support for EMBEDDING_LOOKUPLev Proleev
Fix: 143934742 Test: quantization coupling tests in CTS and VTS Change-Id: I3be053e26686d5cd4a497a4e3390d746aedba758
2019-12-12Add TENSOR_QUANT8_ASYMM_SIGNED support for ROI_ALIGNPrzemyslaw Szczepaniak
Change-Id: I0e765b7c24e83296b22f34f0823d33956db38bab Fix: 143935042 Test: quantization coupling tests in CTS and VTS
2019-12-12Add TENSOR_QUANT8_ASYMM_SIGNED support for SLICEPrzemyslaw Szczepaniak
Change-Id: I44838c1c57db0923b20ae99b1bfd5478152129b9 Fix: 143935043 Test: quantization coupling tests in CTS and VTS
2019-12-12Add TENSOR_QUANT8_ASYMM_SIGNED support for ROI_POOLINGPrzemyslaw Szczepaniak
+ Fix types.spec typos in SIGNED output constrains for few other ops Change-Id: I1531518359b8c450b970425a5be0a7db3d31bd5e Fix: 143934727 Test: quantization coupling tests in CTS and VTS
2019-12-12Merge changes I7f216d62,I7dbf6431TreeHugger Robot
* changes: Add TENSOR_QUANT8_ASYMM_SIGNED support for SOFTMAX Add TENSOR_QUANT8_ASYMM_SIGNED support for FULLY_CONNECTED op
2019-12-12Merge changes I61a76588,Icf8c92bfTreeHugger Robot
* changes: Add TENSOR_QUANT8_ASYMM_SIGNED support for DEPTHWISE_CONV_2D Update DEPTHWISE_CONV_2D to use OperationResolver
2019-12-12Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for DEQUANTIZE"TreeHugger Robot
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for SOFTMAXLev Proleev
Fix: 143935112 Test: quantization coupling tests in CTS and VTS Change-Id: I7f216d6248115ae10b599df0feeeb439b2cbd916
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for DEPTHWISE_CONV_2DLev Proleev
Fix: 143935136 Test: quantization coupling tests in CTS and VTS Change-Id: I61a76588e74332e4acd210d99670269730bf9b4d
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for DEQUANTIZELev Proleev
Fix:143934768 Test: quantization coupling tests in CTS and VTS Change-Id: Ia45ff126fb6d42db80182ffe9ccc1b7269d30287
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for FULLY_CONNECTED opLev Proleev
Fix: 143934556 Test: quantization coupling tests in CTS and VTS Change-Id: I7dbf6431f0492155721755410fd9247d4cb71e3a
2019-12-11Update DEPTHWISE_CONV_2D to use OperationResolverLev Proleev
Bug: 143935136 Test: NNTest_static Change-Id: Icf8c92bf01e547ade01ba3dddbcd9b99d5d1b441
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for (MIN|MAX)IMUMPrzemyslaw Szczepaniak
Change-Id: I027cf2656b23ac551160364ebe8ff1a96fd04321 Fix: 143934186 Fix: 143934169 Test: quantization coupling tests in CTS and VTS
2019-12-11Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for comparison ops"TreeHugger Robot
2019-12-11Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for ARG(MIN|MAX)"Przemyslaw Szczepaniak
2019-12-11Add TENSOR_QUANT8_ASYMM_SIGNED support for comparison opsPrzemyslaw Szczepaniak
Change-Id: I604909df5a7a460df62813656625e4ba48c1de69 Fix: 143934470 Fix: 143934586 Fix: 143934587 Fix: 143934744 Fix: 143935041 Fix: 143935054 Test: quantization coupling tests in CTS and VTS
2019-12-10Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for PRELU"Przemyslaw Szczepaniak
2019-12-10Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for EXPAND_DIMS"Przemyslaw Szczepaniak
2019-12-10Merge "Add TENSOR_QUANT8_ASYMM_SIGNED support for GATHER"Przemyslaw Szczepaniak
2019-12-10Add TENSOR_QUANT8_ASYMM_SIGNED support for GATHERPrzemyslaw Szczepaniak
Change-Id: I7aacc3a6aa5c331f4331b32ed28017515265db7c Fix: 143935114 Test: quantization coupling tests in CTS and VTS
2019-12-10Add TENSOR_QUANT8_ASYMM_SIGNED support for EXPAND_DIMSPrzemyslaw Szczepaniak
Change-Id: I802925b1e990b099b710d82971ad84ea077f66db Fix: 143934168 Test: quantization coupling tests in CTS and VTS
2019-12-10Add TENSOR_QUANT8_ASYMM_SIGNED support for ARG(MIN|MAX)Przemyslaw Szczepaniak
+ Update mutate skips for CAST Change-Id: I82f398d72b234ab4b494a26f0c95f91fe35927b1 Fix: 143934468 Fix: 143934167 Test: quantization coupling tests in CTS and VTS
2019-12-10Merge changes Ic26d37c4,Ieb93203d,I30f46866TreeHugger Robot
* changes: Add TENSOR_QUANT8_ASYMM_SIGNED support for TRANSPOSE_CONV_2D Add TENSOR_QUANT8_ASYMM_SIGNED support for CONV_2D Add TENSOR_QUANT8_ASYMM_SIGNED support for L2_NORMALIZATION
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-22 14:51:16 +0000