| 3c2272e1cf69ef14d1a5dce34541bfd492272f69 |
|
18-Apr-2018 |
Michael Butler <butlermichael@google.com> |
Create _relaxed tests for newly added tests -- generated tests
New tests were added to V1_0 and V1_1 specs, but they did not have a
"_relaxed" test in V1_1.
This CL generates the CTS and VTS tests as a followup CL.
Bug: 76434855
Test: mma
Test: NeuralNetworksTest_static
Test: VtsHalNeuralnetworksV1_0TargetTest
Test: VtsHalNeuralnetworksV1_1TargetTest
Change-Id: Id3703db7091583de19e22b82eab75e88da785428
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 6d78cf0e572d4117b3011bcf0ab768c47dc58031 |
|
12-Apr-2018 |
Mika Raento <mikie@google.com> |
Clarify Fully Connected Op comments and naming
- Make the reshaping of input clearer in header comment
- Make variable names in implementation match header comments
- Add a test for the case where first dimension is not the batch size
Bug: 74851002
Test: NeuralNetworksTest
Change-Id: Ic914d5b61ef162f2be04834bcd128208f13a627f
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| b66b5abcef40e88df6ac18ce7890372192102323 |
|
02-Apr-2018 |
Miao Wang <miaowang@google.com> |
Finalize STRIDED_SLICE
- Add documentation for begin_mask, end_mask, shrink_axis_mask.
- Add support for shrink_axis_mask, and corresponding tests.
- Add validation tests for STRIDED_SLICE.
- Updated existing tests and re-generated CTS and VTS tests.
Bug: 73661082
Test: NeuralNetworksTests_static
Change-Id: I49ad2ffcb3b7496ee1225618833ba324bb01c5b5
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 02e33fc0ef3dce28cee468d1fa4e37511ba1dd48 |
|
23-Mar-2018 |
Miao Wang <miaowang@google.com> |
Make NNAPI support BroadcastSub and BroadcastDiv.
- Make the CPU implementation able to handle BroadcastSub and
BroadcastDiv.
- Add corresponding tests.
Bug: 73661777
Test: mm
Test: NeuralNetworksTest and VTS tests
Change-Id: I0b7921b6f0e9f298e94dcaebe5cfe9ac580295a3
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 3581e149f52fb810ecffe087cfec82cd55ccec22 |
|
26-Feb-2018 |
Miao Wang <miaowang@google.com> |
Generates new V1_0 tests from the newly added specs.
Test: mm
Test: NeuralNetworks tests pass on marlin
Change-Id: I8f964ee6fa090b4fb3c1c6ee7f9a8c1fdac78ae7
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 70f401e3df8435bbb750e77a2341f1feec719556 |
|
23-Feb-2018 |
Miao Wang <miaowang@google.com> |
Add quantized MEAN tests
Bug: 63911257
Test: mm
Test: NeuralNetworksTests and VTS pass on walleye
Change-Id: Ia2374f5678202f1ae6aa2116f3eea1c739abcb50
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| e8e5d34c4159532eb324df393c2c752a508bced1 |
|
23-Feb-2018 |
Miao Wang <miaowang@google.com> |
Generate relaxed precision tests for CTS and VTS.
Bug: 63911257
Test: mm
Test: CTS and VTS pass on walleye
Change-Id: I62655872423e55201e94c05e2cd9da86815d4b26
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 65db490301bb2e35864fb0f2aef67fa8c4be2b88 |
|
20-Feb-2018 |
Miao Wang <miaowang@google.com> |
Add more tests for v1.1 ops
Bug: 63911257
Test: mm
Test: NeuralNetworksTests and VTS tests pass
Change-Id: Ia9674e5bed390fd2507c6c70f1a37737f7733ae9
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 5b0a7d939acba4c70027ea2798e7bbe054e77707 |
|
06-Feb-2018 |
Miao Wang <miaowang@google.com> |
Create NNAPI v1.1 VTS tests.
- Split specs files into corresponding subfolders with versioning.
- Update scripts to be able to handle versioned specs.
- Generate all_generated_$VER_vts_tests.cpp for VTS.
- Generate V1_1 op tests for VTS.
Bug: 63911257
Test: mm
Test: NNAPI VTS tests pass on v1.0 and v1.1 sample devices.
Change-Id: I1980561b69a463f4284f539cc252d298aa99a34c
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 4d8c48b3cc55c0748ef87a0ba8bbbd3959680213 |
|
30-Jan-2018 |
Miao Wang <miaowang@google.com> |
Allow CpuExecutor to execute the new ops.
- Add basic tests for all the newly added operations.
Bug: 63911257
Test: mm
Test: NeuralNetworksTest (CTS) including new tests pass
Change-Id: I78124e5d5dac6824da678342a02792439d1c725e
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 8db6e487763be346310fc5468bd8f33f255ccb9c |
|
06-Oct-2017 |
Yang Ni <yangni@google.com> |
Adding filter rank to SVDF
Bug: 67597181
Pulled in the latest implementation from TF Lite.
This allows a rank parameter that is greater than 1.
Split the state into input and output states.
Added a test spec for rank 2 SVDF.
Test: NeuralNetworksTest
Test: SVDFTest: adb shell /data/nativetest64/svdf_test/svdf_test
Change-Id: I207cc081019369bf5e643e16e91ff664d10975cc
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 4062279c74f2c995349b7ff04f22feb89eb81e47 |
|
19-Oct-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Slicing a given model up to the given first N operations
Also regenerated all CTS, VTS and unit tests, as a test harness function
is renamed.
Test: test.py and sliced mobilenet_quantized up to first 20 ops
Test: unit tests and VTS on Sailfish
Bug: 67781676
Change-Id: I68dcdbda17f8bedc98b1da573a8f884da0d5722d
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 4107f3201f58a9780e2550a83e90b6ec8a80cb1d |
|
27-Oct-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Modernize the test specs of various outdated tests
These generated test were in an earlier format for test_generator,
and were not re-generated when running generate_test.sh
This CL converts the test spec to a modern test spec. Removed manually
written includes in TestGenerated as well.
Also removed test harness for those outdated tests since it is no
longer used.
Bug: 68383630
Test: NeuralNetworksTest on Sailfish
Change-Id: I994317516794f6198436ee5c476cc2f4dbf764da
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| a1ef918e19d3d66773dfbd6b3ccc6bacfd153a52 |
|
27-Oct-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Modernize the test spec of mobilenet_224_gander_basic_fixed
The generated test was in an earlier format for test_generator in which
weights were defined as automatic variables. That is now considered a
bug.
This CL fixes the problem by converting the test spec to a modern test
spec. Removed manually written includes in TestGenerated as well.
Bug: 68383630
Test: NeuralNetworksTest on Sailfish
Change-Id: I4a4325109bb73ab1d63169735e8f1170fe0eaa83
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 9b8ba9d9e398b8448a1e6af568fd1195bc035deb |
|
24-Oct-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Include a full example into mobilenet_quantized spec
Manually merge the generated example file into the spec and convert it
into a properly defined Example. Updated generate_test.sh as the hack to
workaround b/67394906 is no longer needed.
This change is also required by the slicing tool.
Bug: 67781676
Bug: 67394906
Test: NeuralNetworkTest on Sailfish
Change-Id: I95b8718e3dab876eb0bb93a1d40f486c5473017f
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| f6d3074a6ce9d327ac4de25ad8b828fb810ae6ca |
|
12-Oct-2017 |
Yang Ni <yangni@google.com> |
Test LSTM, RNN, and SVDF with non-zero states
Bug: 67427635
These are auto-generated tests. Updates the specs so that states
are not zeros.
Test: NeuralNetworksTest
Change-Id: Id419d020a8c0abc1bf03242c99ad03b2cb192ccb
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 9949232ba3d5b1d95f5b3d9ea310b88b81ee1a45 |
|
07-Oct-2017 |
gfan <gfan@google.com> |
Added one more test spec for space_to_depth/depth_to_space
Bug: 63905942
Test: ran NeuralNetworksTest on marlin
Change-Id: I11df832b5e69479b7fe4e8beff9c0add2ed9992c
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| bee07f73a5f998a2dd6dc581e7776557c21f9684 |
|
09-Oct-2017 |
Miao Wang <miaowang@google.com> |
Force CpuExecutor validating user-provided model output operands.
- For operands with OperandLifeTime::MODEL_OUTPUT, the dimensions,
type, and other meta-data must match the output Shape calculated
from the operation preparation step.
- Fix the ill-defined tests caught by the added validation.
- Incidental changes: generated more tests from tests specs.
Bug: 67390841
Test: NeuralNetworksTests pass
Change-Id: I40d35db0f7a868feae773dbf7e12cf4bf5f5e275
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 66d5cb6e3a90aefc8d545f6369080ab88de9d667 |
|
05-Oct-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Rename setInputsAndOutputs
To avoid confusion with other set* functions, rename setInputsAndOutputs
to identifyInputsAndOutputs.
Also added a few new generated tests when I reran generate_test.sh.
Bug: 63905942
Test: local & VTS tests
Change-Id: Ie947044bee1a97426c847b9d66e60f0c25395bbc
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| a5e146a1efaf31028e2eeef53dd3264fd6d95563 |
|
05-Oct-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Change stateful Ops to stateless ones" into oc-mr1-dev
|
| eefb1e60444afd08a4350e11f281ac7064ebba63 |
|
04-Oct-2017 |
Yang Ni <yangni@google.com> |
Change stateful Ops to stateless ones
Bug: 63905942
Updated Ops RNN, LSTM, and SVDF.
Split outputs used for states into inputs and outputs.
Test: NeuralNetworksTest
Change-Id: Ia3d11f640cba4cab1b94d0b9746c46d347c024a4
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| e68d924a02511200be6186c947359dee7bb58f23 |
|
04-Oct-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Modify test files for size. Add tests.
Modified the specs so that the generated data files won't be
so large as to take very long to compile.
Also generate the tests for the new specs that have been added.
Finally, update the VTS tests for the specs we have.
Bug: 67394766
Test: Compiled and ran unit tests and VTS tests.
The regular tets pass except for:
[ FAILED ] GeneratedTests.conv_quant8_channels
[ FAILED ] GeneratedTests.conv_quant8_large
[ FAILED ] GeneratedTests.conv_quant8_overflow
The VTS tests pass except for:
[ FAILED ] NeuralnetworksHidlTest.conv_quant8_channels
[ FAILED ] NeuralnetworksHidlTest.conv_quant8_large
[ FAILED ] NeuralnetworksHidlTest.conv_quant8_overflow
[ FAILED ] NeuralnetworksHidlTest.lsh_projection
[ FAILED ] NeuralnetworksHidlTest.lstm2
[ FAILED ] NeuralnetworksHidlTest.lstm3
[ FAILED ] NeuralnetworksHidlTest.lstm
Change-Id: I5a26d0d05e1eca9874b0e9b09f33f345b657667a
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 224c01eb06d3dc496b99b0827fdcc9e65bfc4f9a |
|
03-Oct-2017 |
Dong Chen <dongchen@google.com> |
Update generated unit tests and vts tests.
Bug: 63905942
Test: NeuralNetworksTest on Angler
Change-Id: I6a12e8e118bf0855ba6fa49065e054a2336473cd
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 25f0f2dcd7049cd9f8a60133697dff114bbbf00e |
|
28-Sep-2017 |
Miao Wang <miaowang@google.com> |
Allow ADD and MUL taking quantized input.
- Implement the quant8 path for ADD and MUL.
- Add tests for quant8 ADD and MUL, both normal and broadcast version.
- Add validation for quantization parameters, forcing scale >= 0 and
zeroPoint in range [0, 255] for QUANT8_ASYMM type operands.
Bug: 63905942
Test: mm
Test: NeuralNetworksTest pass
Change-Id: Ib643217737af9624a079369e0aea792025223fbb
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| b3ea13a8e7b607d78cceae07a000df073e1520ea |
|
29-Sep-2017 |
Miao Wang <miaowang@google.com> |
Add quantized MobileNet test
Bug: 63905942
Test: mm
Test: NeuralNetworksTest pass
Change-Id: I1abe8074c498da38852f759a3cb7ba6882a7529e
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 1b8fa4fb29e498173f6b7992344a24bce447de82 |
|
14-Sep-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Re-generate CTS tests from test specs
Bug: 63905942
Test: NeuralNetworksTest on Angler
Change-Id: I002efa72d08282ec1a9fa7625f2df4a36dd16fb2
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| bf15cf2352cd159679210c0e575a26e2e37b770a |
|
08-Sep-2017 |
Yang Ni <yangni@google.com> |
Test spec for the Basic RNN operation
Bug: 63905942
Added test spec for RNN.
Additional changes to make the generated test work:
* Updated auto test harness to use EXPECT_NEAR with an absolute
error of 1.0e-5 for comparison of floating point numbers.
* Made the generate example (inputs and golden outputs) local variable
to avoid issues in the second test run.
Test: run generate_test.sh and run the generated tests.
Change-Id: I8844fd134c6718707d63af0dc4b8447ce5257954
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| ec8e1c71d11f7b5abe306e0c9c6af7da755bd46b |
|
13-Sep-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Support ignored output in test generator
Added a new construct IgnoredOutput for specifying outputs that are
excluded from comparing against golden, required by ops like SVDF.
Also regenerated CTS tests since a test now has one more generated
function bool is_ignored(int);
In addition, updated test generator regression suite accordingly.
Bug: 63905942
Test: NeuralNetworkTest on Angler
Test: VtsHalNeuralnetworksV1_0TargetTest on Angler
Test: tools/test_generator/tests/test.py on host
Change-Id: I99a438400c264cd3b4a12014e1cc473a7d324df6
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 9b69c561a47cdb35b911b163c0d819f7a77efe41 |
|
12-Sep-2017 |
Yang Ni <yangni@google.com> |
Fix auto-generated LSH Projection test
Bug: 63905942
Leaving an optional argument unset leads to wrong results.
Setting it an empty python list, i.e., [], fixes the issue.
Test: adb shell /data/nativetest64/NeuralNetworksTest/NeuralNetworksTest
--gtest_filter=GeneratedTests.lsh_projection
Change-Id: I955bb1093b7281711f1fdc4c888aa2891c135730
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 608124c8a4bc3a5385694d74f0b2f2a6ce5d6292 |
|
12-Sep-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Regenerate CTS and unit tests from specs
- Temporarily took out LSH projection as there's a result mismatch
- Added hashtable_lookup and relu1_quant8
Bug: 63905942
Test: NeuralNetworksTests on Angler
Change-Id: Ie3181e39ad062764dc4a37e7d800367c564bef9f
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 62ff6a785f5139658c25124c3af20c489d6fe5f2 |
|
12-Sep-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Regenerate CTS and unit tests from specs
- Temporarily took out LSH projection as there's a result mismatch
Bug: 63905942
Test: NeuralNetworksTests on Angler
Change-Id: Ic8f8281c3d8a785bc8f4d8ea1d307f2d9d402ae5
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 6e4193a640d8610baad9cf88ed4675cfaa4fc874 |
|
11-Sep-2017 |
Dong Chen <dongchen@google.com> |
Fix MAX_POOL quant8 op test breakage and update generated models & examples.
MAX_POOL quant8 op test was broken after explicit padding change in ag/2874247.
This CL fixes it and updates generated models & examples to prevent future
breakages.
Bug: 63905942
Test: NeuralNetworksTest on Angler
Change-Id: I0c63e2b5884f48a5fbe32ebcdff641196b3541b4
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|
| 6a0d306cf902e13ab147c7533b2cb02540ee66d5 |
|
08-Sep-2017 |
I-Jui (Ray) Sung <ijsung@google.com> |
Support example that has tensors of different types
Operations like LSH_PROJECTION requires tensors of different types. This
CL adds support to the generator and harness (for native tests).
Revised:
- Test generator
- Test harness
- Existing examples
Also regenerated examples from testcases.
Bug: 63905942
Test: tools/test_generator/test.py
Test: NeuralNetworkTest on angler
Test: cts-tradefed run commandAndExit cts -m CtsNNAPITestCases
Change-Id: I000cdac92468659709be18599a425ef82695011a
/frameworks/ml/nn/runtime/test/generated/all_generated_tests.cpp
|