summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-06-22Sets bias quant scale in QLSTM test to 0 as it's unused.Viet Dang
Bug: 155421116 Test: NeuralNetworksTest_static Change-Id: I434d70e30b137a37febfab83127d79f9a6ecbae6
2020-06-18Snap for 6601700 from e941d7c787df84718e89ee57e6f71850d539cf6a to rvc-releaseandroid-build-team Robot
Change-Id: I0426ecb5a2d705a9f9c84214ea5c3712171c906b
2020-06-17Merge "Remove unknown rank control flow tests from VTS" into rvc-dev am: ↵Slava Shklyaev
e941d7c787 am: e8a7d0f6a9 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11867240 Change-Id: I0da07295387efb300957b8dbbce1d3d7fb3f9cb5
2020-06-17Merge "Remove unknown rank control flow tests from VTS" into rvc-dev am: ↵Slava Shklyaev
e941d7c787 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11867240 Change-Id: Ib55d105ba54c45b6b79a201829ae82fa65545267
2020-06-17Add control flow tests with unknown dimensionsSlava Shklyaev
Bug: 132458982 Bug: 154597673 Test: NNT_static --gtest_filter='*unknown_*' Change-Id: I514b78ee3cf2efd86a47412e413a6873341a59d0
2020-06-17Merge "Remove unknown rank control flow tests from VTS" into rvc-devSlava Shklyaev
2020-06-17Remove temporary test filters am: 4940d16f03 am: 68870b3878Miao Wang
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11882615 Change-Id: Iffb46dd741e98ccf80b082b70b01ab1520181103
2020-06-17Remove temporary test filters am: 4940d16f03Miao Wang
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11882615 Change-Id: I778f972749a52eff87872f80ba0507ad0c5500ea
2020-06-16Remove temporary test filtersMiao Wang
Bug: 153876253 Test: mm Change-Id: I4d36ed2aeaa7666fa349b50c93397792e34fad12
2020-06-16Remove unknown rank control flow tests from VTSSlava Shklyaev
Fix: 159076604 Test: m Change-Id: I44f31a004309594f03ef7eadf10c2ff51097e954
2020-06-16Snap for 6592941 from e19d742a8f5809a542a096579d0e4830b6e356fe to rvc-releaseandroid-build-team Robot
Change-Id: I6a03d021c58a0264ab9e609ed2fc35fb84c5a2ad
2020-06-15Merge "Change TEST_MAPPING configuration of NeuralNetworksTest_static" into ↵Michael Butler
rvc-dev am: e19d742a8f am: dfa49a7ac6 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11815486 Change-Id: I1b9abbcb8b44ab7be0f1cb9ff475d82a7f84a53f
2020-06-15Merge "Change TEST_MAPPING configuration of NeuralNetworksTest_static" into ↵Michael Butler
rvc-dev am: e19d742a8f Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11815486 Change-Id: Iad612cd6e715e034c0c056869c4c14e5a60a6db2
2020-06-15Merge "Change TEST_MAPPING configuration of NeuralNetworksTest_static" into ↵Michael Butler
rvc-dev
2020-06-12Change TEST_MAPPING configuration of NeuralNetworksTest_staticMichael Butler
This change restricts presubmit to only run a single pass of NeuralNetworksTest_static ("pass 10"), corresponding to: * useCpuOnly = 0 * computeMode = ComputeMode::ASYNC * allowSyncExecHal = 1 Bug: 131770421 Test: mma Test: atest Change-Id: Ifd316ea87151ae80001208a38cb7af321f7cf9ec
2020-06-13Snap for 6586393 from 73ef4105a24f21ec8d6b1bbe4253552e3ba5db26 to rvc-releaseandroid-build-team Robot
Change-Id: I0666b757b0955967dd54948f25450563e7259491
2020-06-12Merge changes I8da1aa92,Ia112e46d into rvc-dev am: 73ef4105a2 am: d2be5141d1Slava Shklyaev
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11680640 Change-Id: I43b1f4cf94266f1fcc6b6359d0f9996cbf7622c9
2020-06-12Merge changes I8da1aa92,Ia112e46d into rvc-dev am: 73ef4105a2Slava Shklyaev
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11680640 Change-Id: I7c255211b7f9d161dfb4afbd52f082cb487a8988
2020-06-12Merge changes I8da1aa92,Ia112e46d into rvc-devSlava Shklyaev
* changes: Support WHILE with growing output tensor in CpuExecutor Relax control flow boundary operand dimension constraint
2020-06-10Support WHILE with growing output tensor in CpuExecutorSlava Shklyaev
1. Adds a WHILE test that produces a tensor of dynamic size. The loop body model has inputs and outputs of unknown rank. 2. Adds an IF test where the branch models have unkown input and output ranks. 3. Modifies the partitioner to avoid trying to delegate referenced models if an IF or WHILE or their associated referenced models have unknown dimensions. 4. Adds dynamic loop output shape support to CpuExecutor by allocating new temporaries on each iteration. Bug: 132458982 Bug: 154597673 Test: NNT_static --gtest_filter='*_dynamic*' Change-Id: I8da1aa92abf090adbdcbf647c113fe27f4c002d8
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
2020-06-10Snap for 6573866 from fe603b440c9d74f9e72c29265f7e208c1b1dea07 to rvc-releaseandroid-build-team Robot
Change-Id: Ic5cf51adcdd15e52785bcafe3d94b241b541e8a7
2020-06-09Merge "Verify non-optional tensors have values in CpuExecutor" into rvc-dev ↵Michael Butler
am: fe603b440c am: bbd60fa020 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11620917 Change-Id: Ibaf84e7080d46cf9e8099aa98480778e5d680f6b
2020-06-09Merge "Verify non-optional tensors have values in CpuExecutor" into rvc-dev ↵Michael Butler
am: fe603b440c Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11620917 Change-Id: I42bebb34dafb55b58a7e329243a539fa15a60f2e
2020-06-09Merge "Verify non-optional tensors have values in CpuExecutor" into rvc-devMichael Butler
2020-06-09Verify non-optional tensors have values in CpuExecutorMichael Butler
This change adds additional validation for non-optional tensors for the following operations: * EMBEDDING_LOOKUP * HASHTABLE_LOOKUP * LSH_PROJECTION * BIDIRECTIONAL_SEQUENCE_LSTM * LSTM * RANDOM_MULTINOMIAL * RNN * SVDF * SPLIT Some operations such as SVDF unpack the scalar values without checking if the value is present, leading to a failed CHECK. This CL adds protections to use default values in these cases, and relies on a corresponding Prepare method to cause these cases to fail validation. Bug: 157516274 Test: mma Test: CtsNNAPITestCases Test: NeuralNetworksTest_static Test: libneuralnetworks_fuzzer Change-Id: I6bb804ec40205c9741b04231022894c714ad28ec
2020-06-04Snap for 6557338 from 54688b80e26325de12e1dd5112a31343f94d68ae to rvc-releaseandroid-build-team Robot
Change-Id: I0285a6d5b4a40655c54c1a0d709b5def5afbf8c8
2020-06-03Relaxed LSH_PROJECTION: do not convert the hash tensor as internal am: ↵Xusong Wang
54688b80e2 am: 7e13bcdcb7 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11707952 Change-Id: Ib8a2e230655ad9c7e3839d79d6a2c3dff6c7f0b5
2020-06-03Relaxed LSH_PROJECTION: do not convert the hash tensor as internal am: ↵Xusong Wang
54688b80e2 Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/ml/+/11707952 Change-Id: I2bed8eabb3110a69fabf6a44a67197af9207bc9e
2020-06-02Relaxed LSH_PROJECTION: do not convert the hash tensor as internalXusong Wang
LSH_PROJECTION is very sensitive to the value of the hash tensor. Prior to this CL, AllInputsAsInternalCoverter will convert the hash tensor to internal by introducing a dummy ADD operation. Under relaxed execution mode, the small precision loss in ADD will result in a significant difference in the final result. This CL prevents the hash tensor from being converted to internal in relaxed precision tests. Additionally, this CL removes a redundant variation in lsh_projection_float16. Fixes: 155962587 Test: NNT_static Change-Id: Id5522b4949a4e3ab4801537e8eb747a25f0cd0e8
2020-05-30Snap for 6542686 from c7e4e604b2d4f1cf8f2bdbe11350a48c719148d3 to rvc-releaseandroid-build-team Robot
Change-Id: Iaaab2d04a5c29c0ddb77949d1ca5e439f45f5329
2020-05-29Merge "Omit dimensions for HAS_NO_VALUE operands" into rvc-dev am: ↵Michael Butler
c7e4e604b2 am: 1a13ed68b3 Change-Id: I574d4545659691bf74623ea2247dd65bbf15339a
2020-05-29Merge "Omit dimensions for HAS_NO_VALUE operands" into rvc-dev am: c7e4e604b2Michael Butler
Change-Id: Ia595deeb58fb6d72e9516082b2c101034a9bde2a
2020-05-29Merge "Omit dimensions for HAS_NO_VALUE operands" into rvc-devMichael Butler
2020-05-28Snap for 6534196 from 2299d26439f28d18aa9d4ef4c6605579ef9bc49a to rvc-releaseandroid-build-team Robot
Change-Id: Id34f07d30e78e2a22c1ccb78a1ba7bb7a6998643
2020-05-27Merge "Use scales representable in fp16 for fp16 casting ops in RGG." into ↵Xusong Wang
rvc-dev am: 2299d26439 am: 469128d85b Change-Id: Idebe9766f3d5c2ce2eac60bdae2c0910e67b0a78
2020-05-27Merge "Use scales representable in fp16 for fp16 casting ops in RGG." into ↵Xusong Wang
rvc-dev am: 2299d26439 Change-Id: I34b0045f0578d6bcca09a72642dfbeed51b74e79
2020-05-27Merge "Use scales representable in fp16 for fp16 casting ops in RGG." into ↵Xusong Wang
rvc-dev
2020-05-27Merge "Add shape check to CAST validation" into rvc-dev am: c3c428e678 am: ↵Lev Proleev
a71f6b8f21 Change-Id: Ib77fd4834e5882fa19c6d8c613a6e2a447aef868
2020-05-27Merge "Add shape check to CAST validation" into rvc-dev am: c3c428e678Lev Proleev
Change-Id: Ic205bc7f63a7d45fe277480d4b09c3cbeb21285b
2020-05-27Merge "Add shape check to CAST validation" into rvc-devLev Proleev
2020-05-27Snap for 6529526 from 66e5923200afc965bf19b880737e9180e9f5c909 to rvc-releaseandroid-build-team Robot
Change-Id: I2607408d3ca618a684ccdaec1c6bcfb4257e1a67
2020-05-26Omit dimensions for HAS_NO_VALUE operandsMichael Butler
Bug: 157268934 Test: mma Test: libneuralnetworks_fuzzer Change-Id: I5d78db36b110eaec2230478c4759f94f386d59c3
2020-05-26Merge changes Ib3b191cc,I9afea607 into rvc-dev am: 66e5923200 am: ad9d2038b2Xusong Wang
Change-Id: Ia6227cb2d15782de915081cdd1e144ee7c938dda
2020-05-26Merge changes Ib3b191cc,I9afea607 into rvc-dev am: 66e5923200Xusong Wang
Change-Id: Ieb84523dbc03cd79cbaf44710e38c2e917d4b96d
2020-05-26Merge changes Ib3b191cc,I9afea607 into rvc-devXusong Wang
* changes: Fix FULLY_CONNECTED issue with unknown num_units. Fix CAST issue with outputs of unknown rank.
2020-05-26Add shape check to CAST validationLev Proleev
The generated test is added only to CTS since VTS would fail on some 1.2 drivers. Fix: 156284111 Test: NeuralNetworksTest_static Change-Id: I4f3c6cdb9f546501e4ca375d7900431c384d6885
2020-05-22Snap for 6520394 from 0824d7c6d6821941bde2d1b82efb7982ff7cc8a4 to rvc-releaseandroid-build-team Robot
Change-Id: I066342232903e34667b03d501433d9823d452570
2020-05-21Use scales representable in fp16 for fp16 casting ops in RGG.Xusong Wang
The scale in NNAPI is always a fp32 data type. As there are accelerators that only support fp16 data type, there is precision loss when executing a casting operation that converts a fp16 buffer to/from a quantized buffer. In the worst case, the precision loss in scale could be up to 0.5 ULP, and results in 0.5 * 255 = 127.5 ULP difference in the final result. This is much larger than the tolerance of these operations in RGG. This CL sets the scale of such operations to be always representable in fp16 to avoid potential precission loss. Additionally, relax the MSE requirement on DEQUANTIZE. Bug: 155842363 Test: NNT_static_fuzzing Change-Id: I511ddb6753be1dabac36e1423820a03a4f28641e
2020-05-21Fix FULLY_CONNECTED issue with unknown num_units.Xusong Wang
Fixes: 156748888 Test: NNT_static Test: 1.3 VTS with ag/11509996 Change-Id: Ib3b191ccefdfd0d03f8c69772976ef0f2421a9d7
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-23 09:08:07 +0000