89e08c0689df686a54292c0e20d7b2bdf8b11b60 |
05-Jun-2018 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Document unspecified dimensions." into pi-dev
|
9ecf6c51674bcf182185d3c80788c5d796415487 |
05-Jun-2018 |
David Gross <dgross@google.com> |
Document unspecified dimensions. Test: mm Bug: 77234888 Bug: 109687058 Change-Id: I8bc8aa1cb3d5092e94f34aa245bc4185516ac434
euralNetworks.h
|
63ca6c1661e50255ca6824e13c49c6e08cdadb5d |
04-Jun-2018 |
Miao Wang <miaowang@google.com> |
Properly guard newly added APIs in P - Use __ANDROID_API__ >= __ANDROID_API_P__ to guard all new APIs added in Android P. Bug: 109661480 Bug: 109661699 Test: mm Change-Id: I3ad4fba01d7950a6a51e6ee3ef7fc96bf9488dc4
euralNetworks.h
|
aec086a4a628f1f186323f7db8f4574515d74f24 |
03-May-2018 |
Michael Butler <butlermichael@google.com> |
Free wrapper resource before assignment Any resources owned by "this" (lefthand side of assignment) are never released (as the destructor would do), but are overwritten by the move-assignment process and are therefore leaked. This CL frees the resource before assignment. Additionally, it adds missing ownership documentation for Compilation. Bug: 79205067 Test: mma Test: NeuralNetworksTest_static Change-Id: Iacd561adee5205cb039fc7dea4fd1a816ca8d641
euralNetworksWrapper.h
|
f1a4694344c8acf2b157724ddf7df6ce2d1ac7bb |
27-Apr-2018 |
Michael Butler <butlermichael@google.com> |
Sync updated NNAPI doc to HAL -- script Additionally reformatted text to fit within 80 character line limit for HIDL HALs. Additionally removed "of each addition" from "Specifies the activation to invoke on the result of each addition" because it was was included regardless of the operation. As one example, it was included in CONV_2D, which (although it includes a bias add) was confusing and unclear. Bug: 72650109 Test: None, only documentation change Change-Id: I779654aa30e1dec1f376ca173257e839fc68dc64
euralNetworks.h
|
eb336294144d6e32ee7e5284cd2c51da32df7ef8 |
20-Apr-2018 |
Michael Butler <butlermichael@google.com> |
Remove inconsistent references to "type" in NNAPI "Type" in the NNAPI refers both to "OperandCode" and "ANeuralNetworksOperandType" throughout the NNAPI documentation. To make the documentation more consistent and less ambiguous, this CL uses documentation links to the name itself instead of "type". Bug: 72650109 Test: None, only documentation changes Change-Id: I6718063da98a9c7ad10c4bb0452bca9a8986270b
euralNetworks.h
|
c8081626b9a9283425430b3775948a64ee305b08 |
26-Apr-2018 |
Mika Raento <mikie@google.com> |
Merge "Script syncing of types.hal from NeuralNetworks.h" into pi-dev
|
3340f8bea2913ee5f4d76193c11f1068b6582770 |
18-Apr-2018 |
Mika Raento <mikie@google.com> |
Script syncing of types.hal from NeuralNetworks.h Automates the synchronization of Operand and Operation enums from NeuralNetworks.h to types.hal Fixed a couple of typos in NeuralNetworks.h that had been fixed in types.hal. Edited NeuralNetworks.h language and formatting to match types.hal for minimal diff and automatic syncing in the future. Change-Id: I527a8829132bf74d4c032056afb9b05d6bb42d04 Bug: 77604249 Test: compared output with git diff Test: mm in $ANDROID_BUILD_TOP
euralNetworks.h
|
2dcdbd97a63fb2563b7fe9865cea8d04481a2e27 |
23-Apr-2018 |
Miao Wang <miaowang@google.com> |
Fix the error code ANEURALNETWORKS_UNMAPPABLE - The error codes should not clash with each other. - Change ANEURALNETWORKS_UNMAPPABLE to be of value 7. - The impact of this enum value change is expected to be minimal. Bug: 68356625 Test: make checkbuild Test: NNAPI CTS and VTS tests pass. Change-Id: I85a8b9a2180659d9057564119802d851e4011fc6
euralNetworks.h
|
a98386dad9f0356e4e51395137d73bd558ed30bf |
27-Mar-2018 |
Yang Ni <yangni@google.com> |
Fixed an error in LSTM documentation on peephole Bug: 76431010 One of the inputs is independent of the other two, and not related to the peephole optimization. Test: None. Documentation change only. Change-Id: Ia8aa3e86f31bfa5f5f8378bb9af1e293eae5bae8
euralNetworks.h
|
369a512b5626af9f1722a9674166adeeed5bc057 |
19-Apr-2018 |
Miao Wang <miaowang@google.com> |
Document that RESIZE_BILINEAR will not force align corner pixels. Bug: 73961830 Test: mm Change-Id: I7c94f3488343db2cb8be6852c08de5cdad6645c9
euralNetworks.h
|
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
euralNetworks.h
|
58cbcf3e037e9f6db040b0f8b4d5fc06d6015e65 |
09-Apr-2018 |
David Gross <dgross@google.com> |
Fix documentation typos. Test: NeuralNetworksTest_static Test: VtsHalNeuralnetworksV1_0TargetTest --hal_service_instance=android.hardware.neuralnetworks@1.0::IDevice/sample-all Test: VtsHalNeuralnetworksV1_1TargetTest --hal_service_instance=android.hardware.neuralnetworks@1.1::IDevice/sample-all Bug: 76016141 Bug: 67601776 (partial fix) Change-Id: Iff881734a2aad5de9b58177111cb9793ea4853a6
euralNetworks.h
|
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
euralNetworks.h
|
56d801b8b91eab50228b754472cf1240a7e39acd |
05-Apr-2018 |
Miao Wang <miaowang@google.com> |
Fix the spec for TENSOR_QUANT8_ASYMM to match our validation. - Scale of 0.0 is invalid for quantized tensor. Bug: 77236592 Test: mm Change-Id: I050d9f4f44134f3193ba4a82b281b2a770a51bfa
euralNetworks.h
|
a9ce0001ca1df32fde3f322c385eef8ce7279329 |
23-Mar-2018 |
Miao Wang <miaowang@google.com> |
Remove unused / unimplemented parameter "crops" for BATCH_TO_SPACE Bug: 73661175 Test: mm Test: NeuralNetworksTests Change-Id: If4b50c50e0847ce36016f6c257aac3b9d418c7b2
euralNetworks.h
|
247996038e3595d638dad6a54aeae69a634d8727 |
22-Mar-2018 |
Michael Butler <butlermichael@google.com> |
Explicitly set dimension to be nullptr when empty vector An empty std::vector is allowed to have any value for the std::vector::data (pointer) field. When a std::vector is transferred into ANeuralNetworksOperandType::dimensionCount and ANeuralNetworksOperandType::dimensions, ANeuralNetworksOperandType::dimensions must be explicit set to nullptr if the std::vector is empty. This CL additionally has come code cleanup, such as removing the now-unused setFromIntList utility function. Bug: 76122637 Test: mma Test: NeuralNetworksTest Change-Id: I097464df78ee037fd87bc8a970f696d5d3bb6e3e
euralNetworksWrapper.h
|
0882987fa68df21b79fd1ddc6538da8a1fde65d2 |
20-Mar-2018 |
Miao Wang <miaowang@google.com> |
Remove obsolete comments about ANeuralNetworksMemory_createShared Bug: 68822426 Test: mm Change-Id: I7546333107281b39de6fbf8ed18371f1ad3dc7cb
euralNetworks.h
|
0deadd8c690b5a131d66e4bc03667315f28df321 |
14-Mar-2018 |
David Gross <dgross@google.com> |
Call validation from ModelBuilder::finish(). Also: - Make incompletely-initialized large operand values recognizable as incompletely-initialized (for easier validation and manual debugging). - Fix typos. Test: nn/runtime/test/NeuralNetworksTest_static Bug: 74784979 Change-Id: I4753ca137228d733b10097e68a3ef3e2fb6f46c7
euralNetworks.h
|
b74d2837ab1687c1a4f913aa5f90a9838efe0add |
21-Feb-2018 |
Miao Wang <miaowang@google.com> |
Add relaxed execution tests - Extend the test generator and test wrapper to allow relaxed execution. - Make the accpeted error range to be 5 ULP of FP16. - Copy and modify existing V1_0 tests to be V1_1 relaxed tests. Bug: 63911257 Test: mm Test: Relaxed tests pass on devices with FP16 support Change-Id: I8954aeeebcf7312d7d94580992708f1b42e2b32e
euralNetworksWrapper.h
|
084401d6215dca122999261c5ac3718ebf61b29e |
14-Feb-2018 |
Michael Butler <butlermichael@google.com> |
NNAPI narrow evaluation for P -- runtime We have determined that for Android P it is sufficient to have a mechanism for a developer to specify on a per-model basis that it is acceptable for FLOAT32 operations to be carried out as if they were FLOAT16 operations. This CL manages the versioning differences between 1.0 and 1.1. Bug: 63911257 Test: mm Test: NeuralNetworksTest Test: VtsHalNeuralnetworksV1_0TargetTest Change-Id: If6f31536eedc92c4795056fdf3ff8818db1bc988
euralNetworks.h
|
58f0467869f275a18ec9c01859fc783ee9933f34 |
29-Jan-2018 |
Miao Wang <miaowang@google.com> |
Add new ops enum to NeuralNetworks.h Test: mm Bug: 63911257 Change-Id: I5e251fd199f3c4114e0275dab3081dc4de384d4e
euralNetworks.h
|
9c683a1fa8c5b8b4860adf7c277df64b38c3a0c9 |
06-Dec-2017 |
Miao Wang <miaowang@google.com> |
Fix a typo for implicit padding doc for DEPTHWISE_CONV_2D Bug: 68318799 Test: mm Change-Id: I216458fe4c1a17ccf1f5dcf20d04ad6f86c9834d
euralNetworks.h
|
a7e5f8d601737eebef06b7b4e3866975cc4645df |
22-Nov-2017 |
David Gross <dgross@google.com> |
Improve error handling for Model::finish(). - finish() fails if Model is in invalid state. - If finish() fails put Model in invalid state. Test: nn/runtime/test Change-Id: I9bf16a332505136d67d371ba5d55f56b7c3cce53
euralNetworksWrapper.h
|
46ada93580e5036924e969b56799e2c62bdb8699 |
01-Nov-2017 |
David Gross <dgross@google.com> |
Add missing Result values to NeuralNetworksWrapper.h. Bug: 68728827 Test: nn/runtime/test Change-Id: I00b4a09b5d0dcf3d1d6776fb4b57c9003820f16f
euralNetworksWrapper.h
|
c308618ad6e6ae365e063bf431db7fa1036eee07 |
30-Sep-2017 |
Yang Ni <yangni@google.com> |
Update LSTM documentation Bug: 67184259 Bug: 67844091 Added the exact equations that are used in the current implementation. Minor clean-ups to activation function, scratch buffer size, and wording about building a model. Local rendering here. http://yangni-z840.mtv.corp.google.com/nnapi/group__NeuralNetworks.html#ggaabbe492c60331b13038e39d4207940e0ad0377e8c305e596fb7f64ff896671fc5 Test: None. Doc change only. Ran doxygen locally and visual inspects. Change-Id: I75548afb126a1c1b30121125fdbc3af9b356b300
euralNetworks.h
|
1af33897ebdc15c41c94f93535df8904065d211b |
14-Oct-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Add documentation for PaddingCode in NeuralNetworks.h Bug: 67331808 Test: ran NeuralNetworksTest on marlin" into oc-mr1-dev
|
d1da6261af7d54c5f8343a18479ab77fe2449f8a |
11-Oct-2017 |
gfan <gfan@google.com> |
Add documentation for PaddingCode in NeuralNetworks.h Bug: 67331808 Test: ran NeuralNetworksTest on marlin Change-Id: I4b879f8a67bc366bffb12601dba7f021659a6921
euralNetworks.h
|
1da8fed77c5c296afa18f754ec3616e7f02a4cfd |
12-Oct-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Pass large model operand values in shared memory. Because of a limitation in HIDL, large tensor values (e.g. weights) need to pass in shared memory rather than as HIDL arguments. This prevented large models from running on a driver. Separated the handling of memory pools so that request and model pools are not comingled. Also improve logging so we see more details about the models when logging. Bug: 67603060 Test: Ran system tests and VTS tests. Change-Id: I760e31275699f9306c4b21945935dc3a4ca68754
euralNetworks.h
|
3a9b7c8eb10692fc746a954bfbc4b53b774d267b |
10-Oct-2017 |
Yang Ni <yangni@google.com> |
Made embedding lookup take int lookups Bug: 67459279 NN API only supported float tensors when this op was first added. Now int tensors are allowed in the API and we have updated the documentation saying the lookup indices are integers, we need update the implementaion and tests of this op to match the documentation. Also reverted the input order to match TF Lite. Test: NeuralNetworksTest Test: adb shell /data/nativetest64/embedding_lookup_test/embedding_lookup_test Change-Id: Ie0db884ca772e8fb388876c3cef609016e753bf1
euralNetworks.h
|
a86baa03c7756a3a3cd9207ef38a23025dd27ad4 |
10-Oct-2017 |
Gerry Fan <gfan@google.com> |
Merge "Minor clarification for NeuralNetworks.h for EMBEDDING_LOOKUP and HASHTABLE_LOOKUP Bug: 67459279 Test: locally run cts NeuralNetworksTests on marlin" into oc-mr1-dev
|
a6ec3b4ac8a993214406235cfabc71f9c10b29b4 |
05-Oct-2017 |
gfan <gfan@google.com> |
Minor clarification for NeuralNetworks.h for EMBEDDING_LOOKUP and HASHTABLE_LOOKUP Bug: 67459279 Test: locally run cts NeuralNetworksTests on marlin Change-Id: Ib0bc48b3d0a2403820989d8c7d517081f5a9b647
euralNetworks.h
|
3a664717ca18db7941130e16538f75e10513a41d |
09-Oct-2017 |
Miao Wang <miaowang@google.com> |
Fix RESIZE_BILINEAR documenation. Bug: 67503562 Test: none needed Change-Id: Ib59ce8d8dbe814b835c400081a286a0514b5f1be
euralNetworks.h
|
7faa4d8a5301bc64e89a0228037a373bc21259c5 |
05-Oct-2017 |
Gerry Fan <gfan@google.com> |
Merge "Fix test spec in depthwise_conv_2d and depthwise_conv_2d_quant8 Bug: 67381597 Test: locally run cts NeuralNetworksTests on marlin" into oc-mr1-dev
|
c96b526122230f552e5bbb96d4bd0497f50c5c91 |
05-Oct-2017 |
gfan <gfan@google.com> |
Fix test spec in depthwise_conv_2d and depthwise_conv_2d_quant8 Bug: 67381597 Test: locally run cts NeuralNetworksTests on marlin Change-Id: Iba23f8d024e87d563b0020b6c02e30b2e65
euralNetworks.h
|
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
euralNetworks.h
euralNetworksWrapper.h
|
4b9f4fa4e9353ec89310041499e82c2bcd1a3090 |
05-Oct-2017 |
David Gross <dgross@google.com> |
Merge changes Ia0ccf4e4,I5767d214 into oc-mr1-dev * changes: Allocate all inter-partition temporaries to a single Memory object. Asynchronus partitioned execution, plus minor code cleanup.
|
a5e146a1efaf31028e2eeef53dd3264fd6d95563 |
05-Oct-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Change stateful Ops to stateless ones" into oc-mr1-dev
|
e3178825b8686f3300a895572691a2e8c1f0676b |
05-Oct-2017 |
David Gross <dgross@google.com> |
Asynchronus partitioned execution, plus minor code cleanup. Partioned execution is now asynchronous end-to-end, although the individual execution steps are serialized with respect to one another. Other changes: - Minor code cleanup per comments from previous CL https://googleplex-android-review.git.corp.google.com/#/c/platform/frameworks/ml/+/3000624/ including comments on memory management for inter-partition temporaries. - Remove "simulation" form of partitioning, and renumber values of debug.nn.partition property accordingly. - Fix android:nn:wrapper::Event -- add missing contructor and fix comment. Bug: 63905942 Test: mma (user) mma (userdebug) ml/nn/runtime/tests (userdebug) (and spot check that debug.nn.partition unspecified, 0, and 1 exhibits the correct control over partitioning behavior) Change-Id: I5767d2144c4a42258444a6445866529919d4fe8d
euralNetworksWrapper.h
|
35d33fbbbda5d379d26a1d414a8e3e016e026938 |
05-Oct-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Revert "Fix a typo in documentation to FULLY_CONNECTED"" into oc-mr1-dev
|
ce38590c9b8e3f8b90c283472b54b47ad7a1f2b9 |
04-Oct-2017 |
Yang Ni <yangni@google.com> |
Revert "Fix a typo in documentation to FULLY_CONNECTED" Bug: 63905942 The apostrophe to the matrix indicates a transpose. This reverts commit 76354c87e4c1aad5da714828a633fe37404920a8. Change-Id: I8e84a2e39e826c8bfe25366cc6d8d1fd64252b4f
euralNetworks.h
|
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
euralNetworks.h
|
2a2b3842abe8bad7cde8a1c1ef5645291fd587b5 |
04-Oct-2017 |
Miao Wang <miaowang@google.com> |
Fix documentation of CONCATENATION. Bug: 67185615 Test: none needed Change-Id: I7100a0f67a80d83c47346d420c97ac83ff31613b
euralNetworks.h
|
a3bed7887a65e0662080d2630e228954ca578ca3 |
04-Oct-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Fix problem passing null optional operands." into oc-mr1-dev
|
76354c87e4c1aad5da714828a633fe37404920a8 |
03-Oct-2017 |
Yang Ni <yangni@google.com> |
Fix a typo in documentation to FULLY_CONNECTED Bug: 63905942 Test: none. Not a code change. Change-Id: I5e03e63a6599355ba970330ca1ad744e6999f695
euralNetworks.h
|
62cc2758c1c2d303861e209f26bddcf4d7564b73 |
28-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Fix problem passing null optional operands. Allow optional operands to be specified in the model as having no value. Modify the HAL to allow passing missing operands. Change the validation of most operations to verify that none of their operands are missing. Bug: 63905942 Test: Compiled and ran runtime/test. Change-Id: I378bc51847a8fb30cb2d36c9bf707ebadc79289b
euralNetworks.h
|
35647da7af1f99b6abb48c4eeaec042ea9edfb4d |
03-Oct-2017 |
Miao Wang <miaowang@google.com> |
Fix documentation of FULLY_CONNECTED, and add checks for input rank. Bug: 67319070 Bug: 63905942 Test: mm Test: NeuralNetworksTests pass Change-Id: Ia19c643598c48682b65d4de984fb22d9c3bb05f9
euralNetworks.h
|
93e086fbc0f1577eabdffd0d3420589f2788bd95 |
02-Oct-2017 |
Miao Wang <miaowang@google.com> |
Remove activation parameter from CONCATENATION Bug: 67185615 Test: mm Change-Id: I3555259305761691ea65d2fff406d7fcd8ee8eb5
euralNetworks.h
|
8e86dabb2a154ba15f519138be46bffc0d30b20c |
30-Sep-2017 |
Yang Ni <yangni@google.com> |
Merge "Updated data type for keys in Hashtable Lookup" into oc-mr1-dev
|
850d1eb131d62b07438ff5869bdb80e09fada766 |
30-Sep-2017 |
Miao Wang <miaowang@google.com> |
Allow implicit padding - Upadte the documentation to support implicit padding. - Removed error logging for implicit padding input. Bug: 63905942 Test: mm Test: NeuralNetworksTest pass Change-Id: I5149433ccb393ed390ad9e24013249ef37c6aba8
euralNetworks.h
|
7da2c9a54b25fd15546d6f99eba06dff1ac9a862 |
29-Sep-2017 |
Yang Ni <yangni@google.com> |
Updated data type for keys in Hashtable Lookup Bug: 63905942 This OP was initially added when the NN API only allowed floating- point tensors. Now since integer tensors are allowed in NN API, update the Hashtable Lookup with the better suited data types. Test: NeuralNetworksTest Change-Id: I3dafb2f60bd428700a5a6de393b86dbd75f285dd
euralNetworks.h
|
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
euralNetworks.h
|
c7e8396248b9f42a5cec48b4e7c1c03f3ba081b4 |
28-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Fix documentation for CONV_2D, DEPTHWISE_CONV_2D, FULLY_CONNECTED." into oc-mr1-dev
|
c6973dfadcfef42729c5e379b149bc332715819f |
28-Sep-2017 |
Miao Wang <miaowang@google.com> |
Fix documentation for CONV_2D, DEPTHWISE_CONV_2D, FULLY_CONNECTED. Bug: 63905942 Test: mm Change-Id: I9d5d47691c0f233d617a1bd4b2dde804a95de343
euralNetworks.h
|
ef22aa5727b96e9a0863ef71cfbe3dbdac339408 |
16-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Create mutiple sample test drivers Create multiple sample test drivers to better test partitioning. Bug: 63905942 Test: Compiled, ran runtimte/tests and VTS tests. Change-Id: I037aeb82a849592f3a931bcdc558f32010aa1146
euralNetworks.h
|
45bf79e5b9fee354fde7c1f64417d9ca4a1da7da |
25-Sep-2017 |
Miao Wang <miaowang@google.com> |
Clarify the expectation of scale and zeroPoint for affected ops. - Remove the OperandType constructor taking f_min and f_max as quantization parameters as it causes confusions for ops like LOGISTIC and SOFTMAX. - Update the documenation to clearly state the expected scale and zeroPoint for LOGISTIC, SOFTMAX and CONCATENATION. - Update the tests to directly input scale and zeroPoint. Bug: 63905942 Test: mm Test: NeuralNetworksTest pass Change-Id: Ia450d6ce9509205d22e6383bd7e454afa0568cbb
euralNetworks.h
euralNetworksWrapper.h
|
3309c4645147c763b3437f7d9d69a8725dc7cd38 |
24-Sep-2017 |
Miao Wang <miaowang@google.com> |
Fix documentation to use zeroPoint instead of offset or zero_value Bug: 63905942 Test: mm Change-Id: I29350d80b7fff597601837870cded41788037cf3
euralNetworks.h
|
874d039215516aebdaba2e242609199897fe80c0 |
23-Sep-2017 |
Miao Wang <miaowang@google.com> |
Fix sigmoid and softmax tests and implementation. - CPU executor now checks both the scale and offset for the output Shape. - The golden references and output range for the tests are updated. Bug: 63905942 Test: mm Test: NeuralNetworksTest pass Change-Id: I9e892ae0de8ea17298dbb7edb96036e1d30c84fb
euralNetworks.h
|
6dd108cb66fa0f79d416428fccd5bc74c4360f1a |
22-Sep-2017 |
Quddus Chong <quddusc@google.com> |
Fixed small typos in header comments. Test: none Bug: 63905942 Change-Id: I879ec3cb5a7b4d62851e286997cdce6dac6924b1
euralNetworks.h
|
dec08deb44c829eca478214972170606323b66cf |
18-Sep-2017 |
Miao Wang <miaowang@google.com> |
Fix documentation of depthwise convolution Bug: 63905942 Test: mm Change-Id: Ied08b698b1c10e4fc0f31445ae50a22c9e3dc8f5
euralNetworks.h
|
67573d1086d40fb50e6fe1dbcb84be43d9f05ac2 |
16-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Fix RNN documentation" into oc-mr1-dev
|
80e4de129633e61d116ddef412beb7e4db93106d |
15-Sep-2017 |
Yang Ni <yangni@google.com> |
Fix RNN documentation Bug: 63905942 The last two inputs were in the wrong order for the RNN operator. Test: ran NeuralNetworksTest on marlin Change-Id: I3e045eddb40409f4beb9f21d266c84d18c5406cb
euralNetworks.h
|
fbc04728deedbbf1a119460f5d3c950328b1aeae |
14-Sep-2017 |
Quddus Chong <quddusc@google.com> |
Clean up docs. Replace these references: 1) ANeuralNetworksExecution_wait with: ANeuralNetworksEvent_wait 2) ANeuralNetworksExecution_start with: ANeuralNetworksExecution_startCompute 3) ANeuralNetworksRequest_free with: ANeuralNetworksExecution_free Test: none Bug: 63905942 Change-Id: Ic8f0631f6e8be894c49d18448538583225a0ddec
euralNetworks.h
|
39dcb0f2cc87b0eba06b314a91d55e50efea5abe |
14-Sep-2017 |
David Gross <dgross@google.com> |
Merge changes I5922f13a,I7194e277 into oc-mr1-dev * changes: Bring back Event, and wait on Event rather than on Execution. Make Compilation API synchronous.
|
425b2594c76e934dfdbc93209253e3c189571149 |
14-Sep-2017 |
David Gross <dgross@google.com> |
Bring back Event, and wait on Event rather than on Execution. This is at the recommendation of the API council. This undoes change I50b8e155d9589210e14920237878a4f295e7755f Bug: 63905942 Test: nn/runtime/test Change-Id: I5922f13ae7070f7ec0af035557a9921c88ca9221
euralNetworks.h
euralNetworksWrapper.h
|
65aa556323f4a054f80a75b6c4c721b2a7ed3298 |
14-Sep-2017 |
David Gross <dgross@google.com> |
Make Compilation API synchronous. Per advice from the API council, the Compilation API no longer exposes asynchronous behavior to the user -- rather than start() and wait() APIs, we now have a finish() API. Also: - Track whether or not Compilation::finish() has been invoked, for the sake of API usage checking. - Changed Model::finish() to return Result not int, for consistency. - Made Model and Compilation documentation more similar to one another. - Fixed typos in documentation. NOTE: It's possible to implement compilation asynchronously behind the synchronous API -- e.g., Compilation_finish() starts compilation asynchronously, and we implicitly wait for it to complete at the point of Request_startCompute(). However, this would defeat one of the purposes of an explicit Compilation API -- giving the application control over when to spend the time compiling, to keep it off a critical timing path, such as repeatedly applying a model to a sequence of frames in real time. Bug: 63905942 Test: nn/runtime/tests Change-Id: I7194e277aa6211af66794d917dcc5254c7d81af4
euralNetworks.h
euralNetworksWrapper.h
|
a5c16e4e73a34e6776a8f071ab884444c28b6893 |
14-Sep-2017 |
Miao Wang <miaowang@google.com> |
Rename ANeuralNetworksOperandType.offset to zeroPoint Bug: 63905942 Test: NeuralNetworksTest Change-Id: I15464657dc4eed9adbfc444df306cd0062993be3
euralNetworks.h
euralNetworksWrapper.h
|
7612f29b31f97f3b15769264131566b36dea9a25 |
12-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Remove the initialization and shutdown APIs. Initialize the list of drivers as needed rather than requiring an API call. This initialization will be triggered when compiling the first model. Bug: 63905942 Test: Compiled and ran the unit tests. Change-Id: I1893a9f9d96d9d241916de3a7dd3cb7a54e33b06
euralNetworks.h
euralNetworksWrapper.h
|
d2d0c031c43e8e5aafc75e8a652d79bcc2aaca99 |
12-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Update documentation. Remove IntList. As NDK council request, updated some documentation and removed IntList. Other issues will be addressed in other CLs. Bug: 63905942 Test: Compiled and ran the unit tests. Change-Id: Iff6898737696daa6bbf41643b07d0a4508af4c0d
euralNetworks.h
euralNetworksWrapper.h
|
66d56404cdfab9ab8aa79d4bda83be3832a3aff9 |
14-Sep-2017 |
Miao Wang <miaowang@google.com> |
Make the enums pass as int32_t, and make offset and length size_t. - Also removes fp16 perf and bootup time info. Bug: 63905942 Test: NeuralNetworksTest Change-Id: I1f722fed06a4227eb5f204eb51f7728fe749f5ec
euralNetworks.h
euralNetworksWrapper.h
|
01df5047f2661fd8428d5fc966bc3b7984ec1857 |
14-Sep-2017 |
David Gross <dgross@google.com> |
Merge "Rename Request to Execution at the API level." into oc-mr1-dev
|
5e7827e7bdc7ee6b6fc9b1989070c98fa10e9797 |
13-Sep-2017 |
David Gross <dgross@google.com> |
Move OEM codes out of NeuralNetworks.h to new file NeuralNetworksOEM.h. Also remove FAKE_QUANT operation. Bug: 63905942 Test: nn/runtime/tests, vts Change-Id: Ia9262ec727f93dfe1b0ccade4ed76347c55ba163
euralNetworks.h
euralNetworksOEM.h
euralNetworksWrapper.h
|
3ced3cfd5b8f22b632c35f24e585c4847383b195 |
13-Sep-2017 |
David Gross <dgross@google.com> |
Rename Request to Execution at the API level. There's a separate notion of Request at the HIDL level; this is not renamed. Bug: 63905942 Test: nn/runtime/tests Change-Id: I671e3b387d735a9091dc5a2d14afbea6b45f444c
euralNetworks.h
euralNetworksWrapper.h
|
05292dc440641a38fab6a07c1372ce6780816c1c |
13-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Update ANeuralNetworksMemory related API" into oc-mr1-dev
|
aca6ab57e0e07367724934dd372e0bf4e8bc6b3c |
13-Sep-2017 |
Michael Butler <butlermichael@google.com> |
Merge "Neuralnetworks HAL cleanup -- frameworks/ml/nn" into oc-mr1-dev
|
5f916fc9a7ae95f172492bdfe5344c37beff3a6f |
12-Sep-2017 |
Michael Butler <butlermichael@google.com> |
Neuralnetworks HAL cleanup -- frameworks/ml/nn Does three primary cleanups: 1) Removes unused OperandTypes, creates OEM types 2) Creates explicit ErrorStatus; functions return status 3) IDevice::initialize renamed to getCapabilities, IDevice::getSupportedSubgraph renamed to getSupportedOperations It makes the corresponding changes to the runtime. Bug: 63905942 Test: mm, frameworks/ml/nn/runtime/test Change-Id: I1545373dfa30780aaf9b61b3938be536cf122e74
euralNetworks.h
euralNetworksWrapper.h
|
42dc6a6cd68877cd85e3bc475b41bda0fd946c41 |
13-Sep-2017 |
Miao Wang <miaowang@google.com> |
Update ANeuralNetworksMemory related API - Remove ANeuralNetworksMemory_createShared. - Remove ANeuralNetworksMemory_getPointer. - Allow a offset specified for ANeuralNetworksMemory_createFromFd. - Update the tests. Bug: 63905942 Test: mm Test: updated NeuralNetworksTests pass. Change-Id: I1fd7362ec379f5f855c71c6758e0b647be9aaa76
euralNetworks.h
euralNetworksWrapper.h
|
b5474e8bbbf5bca2cf93d6c2bda1b56981428bad |
13-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Documentation on speech/text OPs" into oc-mr1-dev
|
0408f166b02c866f141034615370f503b428c71a |
11-Sep-2017 |
Yang Ni <yangni@google.com> |
Documentation on speech/text OPs Bug: 63905942 Operations documented in this CL include: Embedding Lookup; Hashtable Lookup; LSH Projection; LSTM; RNN; SVDF. Test: mm Change-Id: I312a5164f63f5f646d7baddb3870e230e008b28a
euralNetworks.h
|
4ffc65b9cb9c9d04c694f0948117554526a46b17 |
12-Sep-2017 |
David Gross <dgross@google.com> |
Remove Event and wait on Request instead. Bug: 63905942 Test: nn/runtime/tests Change-Id: I50b8e155d9589210e14920237878a4f295e7755f
euralNetworks.h
euralNetworksWrapper.h
|
133ad4f3efd94aefc47d137e12b16e0b41cd6c4a |
11-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Cleaned up the documentation. - Fixed many formatting issues that would give Doxygen trouble. - Corrected information on multithreading. - Corrected misc. errors. - Removed remaining TODOs. Bug: 63905942 Test: Compiled & ran the unit tests. Change-Id: I9c3f62b27fec372ee17d99323a129cbb9c3acbe1
euralNetworks.h
|
18c58d289c2346d750301392866229630960b392 |
12-Sep-2017 |
David Gross <dgross@google.com> |
Merge "Fix bug in the way an Event handles a bound thread." into oc-mr1-dev
|
d756f33874dccaedd99eef78942eaf4ae2389ed9 |
12-Sep-2017 |
David Gross <dgross@google.com> |
Merge "More discipline for models and requests." into oc-mr1-dev
|
402baa3fde7795bcea6e2ebeffab97c70b228dd4 |
12-Sep-2017 |
David Gross <dgross@google.com> |
Fix bug in the way an Event handles a bound thread. Previously, we were calling std::thread::join() from the Event destructor. However, if the Event's reference count drops to zero due to action in the bound thread, then the bound thread calls the Event's destructor, and tries to join itself; this throws an exception, and produces a message like terminating with uncaught exception of type std::__1::system_error: thread::join failed: Resource deadlock would occur However, we do not seem to have had any test cases that actully call ANeuralNetworksEvent_free(), so every Event would leak, and we never hit this problem. Now also change Request::compute() to stop leaking its event -- doing this without changing bound thread handling would produce the exception shown above. Bug: 63905942 Test: nn/runtime/tests Change-Id: I0d9370d187c10be00a4a69db8eb5f79854c9be12
euralNetworksWrapper.h
|
7c9cca3ae67378313c586e010f32fb29e648da95 |
12-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Add documentation for ops" into oc-mr1-dev
|
544739620cd7f37d40524d2407c92042e485c73f |
11-Sep-2017 |
David Gross <dgross@google.com> |
More discipline for models and requests. - Must explicitly call new finish() API on model before compiling or freeing. - Must not modify a request once start() has been called on it. Bug: 63905942 Test: nn/runtime/tests, nn/common/operations tests Change-Id: Ifc6e614bda647d729e8702023a02613e629ca6a0
euralNetworks.h
euralNetworksWrapper.h
|
85e34644847dfc0c210520a21217aeb6269f3d3d |
10-Sep-2017 |
Miao Wang <miaowang@google.com> |
Add documentation for ops This CL adds/updates documentation for the following ops: ADD AVERAGE_POOL CONCATENATION CONV_2D DEPTHWISE_CONV_2D DEPTH_TO_SPACE DEQUANTIZE FLOOR FULLY_CONNECTED L2_NORMALIZATION L2_POOL_2D LOCAL_RESPONSE_NORMALIZATION LOGISTIC MAX_POOL_2D MUL RELU RELU1 RELU6 RESHAPE RESIZE_BILINEAR SOFTMAX SPACE_TO_DEPTH TANH Bug: 63905942 Test: mm Change-Id: I23818067bebd1c510ed08b9ac466b7cfb28de039
euralNetworks.h
|
83e24dc4706a5b7089881a55daf05b3924fab3b7 |
10-Sep-2017 |
David Gross <dgross@google.com> |
Add compilation APIs. For now, compilation is implemented as a synchronous no-op. Bug: 63905942 Test: nn/runtime/tests, nn/common/operations tests Change-Id: I1cb96a7e800671d0c70c9cc1b7eab6f8244232cb
euralNetworks.h
euralNetworksWrapper.h
|
23eda93ed8fd41bb4b35359ca91bf8163e20bccc |
09-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Explicitly add "_2D" to applicable ops, and add fused activation function enum." into oc-mr1-dev
|
55cef2193279a646292fdeb355399627dd8990c8 |
09-Sep-2017 |
Miao Wang <miaowang@google.com> |
Explicitly add "_2D" to applicable ops, and add fused activation function enum. - Add the fused activation enum to explicitly set the expectation that only the included ones are supported for fused ops. - Add "_2D" to CONV, DEPTHWISE_CONV, AVG_POOL, MAX_POOL, and L2_POOL, since they are only supposed to do 2D computation. - Updated the tests and the stack to plumbing the changes through. Bug: 63905942 Test: mm Test: NeuralNetworksTests pass for both CPU path and sample hidl driver. Change-Id: I60aeabab1d193d26d938c0d1d3ffdd519740d222
euralNetworks.h
|
0afe5897f4034528b027294efbe45c836924643c |
09-Sep-2017 |
Michael Butler <butlermichael@google.com> |
Merge "Disallow copy semantics for NeuralNetworkWrapper objects." into oc-mr1-dev
|
fce2d202e5dd6f4b2193d84d8726ff4ba86e2fc8 |
09-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Add support for mmaped file descriptor" into oc-mr1-dev
|
105807d963d969197fe78185ed588bfad3dc0ea5 |
05-Sep-2017 |
Miao Wang <miaowang@google.com> |
Add support for mmaped file descriptor - create new API to create ANeuralNetworks_Memory from a mmaped fd. - add the corresponding Memory constructor to NeuralNetworksWrapper.h - Add unit test TestFd for the new API. Bug: 63905942 Test: mm Test: TestFd pass with CPU path. Test: TestFd pass on sample HIDL driver, with ag/2588539 Change-Id: I1fcb0f0bd01e58bd592e18eff11c27034ee4a4c3
euralNetworks.h
euralNetworksWrapper.h
|
cc7f99f85cf9ee7df37e22ed5c957c76713b227d |
07-Sep-2017 |
David Gross <dgross@google.com> |
Remove stray uses of "shared" from comments. Bug: 63905942 Test: mm Change-Id: I1e56bc8b95566c849b80982fa8a7c7ae95030d04
euralNetworks.h
|
7b87fec6bb919d16a8cc2820d470733a2776e8fa |
07-Sep-2017 |
Michael Butler <butlermichael@google.com> |
Disallow copy semantics for NeuralNetworkWrapper objects. Before this CL, the NeuralNetworks runtime would double-free a NeuralNetworks runtime object if its corresponding wrapper object was copied as in the following scenario: { Model base_model = Model(...); // ANeuralNetworksModel_create x1 Model copied_model = base_model; // Copies runtime object's pointer } // ANeuralNetworksModel_free x2, double-free This CL removes copy semantics on Wrapper objects. Additionally, it explicitly handles move semantics to ensure the runtime object is only freed once. Bug: 63905942 Test: mm -j40 Change-Id: Ibd0e8e67eb851a97c18482683c8a37edd665b162
euralNetworksWrapper.h
|
04244491820e2e3c3bbe228948aee7baa11b65bd |
04-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Improvements to the Memory code." into oc-mr1-dev
|
2150f1d186b2854fb5aa609594be12a667f845f0 |
01-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Improvements to the Memory code. Addresses the comments of a previous CL that arrived after submission. Change_Id of that CL was I997a10f6102dcae7c6876c96a4f6f913eaf54c74. - Improved documentation. - getPointer is now getPointerAndSize. It returns a status code, as not all Memory will be able to return a pointer. Test: Compiled and ran the unit test. Bug: 63905942 Change-Id: I8aade863212dd7e95856d7c3d583c1b5f00eda48
euralNetworks.h
euralNetworksWrapper.h
|
689d892203c06c66c7bb2e374462a8434e40b75f |
01-Sep-2017 |
Michael Butler <butlermichael@google.com> |
NNAPI Concurrent Query Management -- Implementation The NNAPI requires requests on a model to be asynchronously processed. This CL implements a basic Event that can later be used to block the runtime thread until the asynchronous request has completed. The design document for NN API asynchronous behavior: https://docs.google.com/a/google.com/document/d/1mO35KK3Mnr489ZftTDnKXXnXiYxk19jZ1C4DWOqaVB4/edit?usp=sharing Bug: 63905942 Test: VtsHalNeuralnetworksV1_0TargetTest (32-bit, 64-bit) with sample driver enabled by cherry-pick frameworks/ml/nn/runtime/test with and without sample driver enabled Change-Id: I97b1d4cbf189176fb3b21b2cc1af09dddaff18ab
euralNetworks.h
|
b14b32a244f4fd6f5b192479c1b3ebbf3999bcff |
01-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Cleanup the constants in NeuralNetworks.h" into oc-mr1-dev
|
c1ab6fc5e953a61811e9f25d2afff06bfa571f0e |
01-Sep-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Cleanup the constants in NeuralNetworks.h - Alphabetize the operations. - Move the constants that count operation and types out of the public header file. - Remove N_GRAM operation. - Add TENSOR_INT32. - Add static assert verifications. - Add documentation. Test: Compiled and ran the tests. Bug: 63905942 Change-Id: Ifafdcb3fd00efabd2b1972b035cc5d43f7dbec51
euralNetworks.h
|
fee08c736c9f273ca29e2dd3d4190121f0276c6e |
01-Sep-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Fix the OperandType constructor in NN wrapper." into oc-mr1-dev
|
910c9f04913e3bee1a0b6406b6e146457d19c5e7 |
31-Aug-2017 |
Miao Wang <miaowang@google.com> |
Fix the OperandType constructor in NN wrapper. - Use constructor delegation in OperandType constructors for the quantized variants. Bug: 63905942 Test: mm Test: Created a quantized AVERAGE_POOL tests, and it pass Change-Id: I1c0a4c26fc056873ff837ba2f60ec85923985fe2
euralNetworksWrapper.h
|
a67d8a0e28baef2fab57a2a3255461869abed74d |
30-Aug-2017 |
Miao Wang <miaowang@google.com> |
Add OEM_OPERATION enum. Bug: 63905942 Test: mm Test: All unit tests passing on sailfish Change-Id: I391ca039620be6525d95ad581acdb6e11d091af2
euralNetworks.h
|
15b59e721efe1a75b0e8cde8bd44bbb532446e13 |
12-Aug-2017 |
Yang Ni <yangni@google.com> |
Implement EMBEDDING_LOOKUP and HASHTABLE_LOOKUP Bug: 63905942 Adapted previous TF Lite implementation and unit tests for NN API on CPU. Test: adb shell /data/nativetest64/embedding_lookup_test/embedding_lookup_test Test: adb shell /data/nativetest64/hashtable_lookup_test/hashtable_lookup_test Change-Id: I6e4a9d8c5d0d47e66e3fa771967da3526d19ac4b
euralNetworks.h
|
b8810af67d4efbd966d6788c56cd01c5aa06a71e |
30-Aug-2017 |
TreeHugger Robot <treehugger-gerrit@google.com> |
Merge "Add full support for Memory pools." into oc-mr1-dev
|
8b99bb1d98a42b67ba1c00e12c7abb3708cf7c05 |
21-Aug-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Add full support for Memory pools. Weights can now be passed via memory pools. Test: Compiled and ran the unit test. Bug: 63905942 Change-Id: I997a10f6102dcae7c6876c96a4f6f913eaf54c74
euralNetworks.h
euralNetworksWrapper.h
|
ca0ae89800ff957f76345c94eeb06d8c5643974b |
29-Aug-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Remove baseline models Also remove submodels, which were to be used with baseline models. Bug: 63905942 Test: Compiled and ran the tests on Marlin. Change-Id: I54495e4e0dfbd6dabf049c8bdc2997600451b047
euralNetworks.h
euralNetworksWrapper.h
|
27e9be3904b034e422ee9b6ab70b35ea994d2b39 |
03-Aug-2017 |
Miao Wang <miaowang@google.com> |
Initial implementation of the following quantized ops. - CONV_QUANT8 - DEPTHWISE_CONV_QUANT8 - AVERAGE_POOL_QUANT8 - MAX_POOL_QUANT8 - LOGISTIC_QUANT8 Additionally, added functions to plumb through quantization parameters. Bug: 63905942 Test: mm Test: end-to-end MobileNet quantized test pass Change-Id: Ib2753c68bf2c51467ae1c158b45541bcfdf10789
euralNetworks.h
euralNetworksWrapper.h
|
707dbd2d55f5dacf78ffb3ad7c8b3f37c2e9d758 |
25-Jul-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
Pass the model as an HIDL component. Major rework where instead of serializing the model and passing it in a shared memory, we pass it via HIDL. The runtime/test code does two run, one goes through the default CPU path, the other through any available driver. Note: The code in hardware/../sample was written by Michael Buttler. Bug: 63905942 Test: Runs both the CPU path and through the sample driver. Change-Id: Ie3ee975d33056ba299895b13193f4698a690dd04
euralNetworks.h
euralNetworksWrapper.h
|
96775128e3bcfdc5be51b62edc50309c83861fe8 |
12-Jul-2017 |
Jean-Luc Brouillet <jeanluc@google.com> |
First implementation of the Neural Networks API. This first version can run a simple query on the CPU either via the fallback path or through a simulated driver. This code has many deficiencies: single threaded, not all validation are done, not going through HIDL, and not enough unit tests. Expect more changes! Test: Compiled and ran the unit tests Change-Id: I9f6a485a2e7207aeb5f91a2904dcb4b7fd8a6f65
euralNetworks.h
euralNetworksWrapper.h
|