- 02 3月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 01 3月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 23 2月, 2022 1 次提交
-
-
由 rockyrhodes 提交于
-
- 22 2月, 2022 1 次提交
-
-
由 dependabot[bot] 提交于
Bumps [tensorflow](https://github.com/tensorflow/tensorflow) from 2.5.2 to 2.5.3. - [Release notes](https://github.com/tensorflow/tensorflow/releases) - [Changelog](https://github.com/tensorflow/tensorflow/blob/master/RELEASE.md) - [Commits](https://github.com/tensorflow/tensorflow/compare/v2.5.2...v2.5.3) --- updated-dependencies: - dependency-name: tensorflow dependency-type: direct:production ... Signed-off-by: Ndependabot[bot] <support@github.com> Co-authored-by: Ndependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
-
- 20 2月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 19 2月, 2022 2 次提交
- 18 2月, 2022 2 次提交
-
-
由 deqiangc 提交于
This PR: 1. Update the tflite_files.txt with two additional files to be pulled in 2. Add build rules to include context_util, builtin_op_data.h and builtin_ops.h 3. Sync from upstream manually BUG=observe errors in automatically sync from upstream Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 deqiangc 提交于
This corresonds to internal cl/429336602 TFLM is replacing GetInput/GetOutput api with more informative micro_context-> AllocateTempXxx/DeallocateTempXxx api for better readability and better temporary buffer auditing. See go/tflm-pairing-allocatetemp-with-deallocatetemp for more info BUG=http://b/217818824Co-authored-by: NTing Yan <94130036+tingyan19@users.noreply.github.com>
-
- 17 2月, 2022 4 次提交
-
-
由 TFLM-bot 提交于
Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 deqiangc 提交于
common.c is renamed to common.cc in upstream github.com/tensorflow/tensorflow/commit/848e84c5a3acda31847457cbd52742436e024984 Hence, change the name to sync it properly BUG=observe error in sync from tf upstream
-
由 Ting Yan 提交于
-
由 Shlomi Regev 提交于
exist) BUG=217301557
-
- 16 2月, 2022 4 次提交
-
-
由 Advait Jain 提交于
BUG=#407
-
由 Ting Yan 提交于
* Add tensor_utils header files from TensorFlow Lite * Removed tensor_utils_common.h and added new file lstm_cell.h tensor_utils_common.h is absorbed into portable_tensor_utils.h in http://cl/428845263 and lstm_cell.h is separated from reference_ops.h in http://cl/428639888 and refactored in http://cl/428808334 for TFLM.
-
由 deqiangc 提交于
* Support building custom ops This is backport of internal cl/24778552 in order to support building custom op from tflite_micro by default. BUG=https://b/278725942 * Remove empty changes Co-authored-by: NTing Yan <94130036+tingyan19@users.noreply.github.com>
-
由 Måns Nilsson 提交于
Fully connected handles more than two outputs. Change-Id: Id8f22cc7d455618926a0f16af61e734d5890e9bb
-
- 15 2月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 12 2月, 2022 2 次提交
-
-
由 TFLM-bot 提交于
Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 TFLM-bot 提交于
-
- 11 2月, 2022 2 次提交
-
-
由 Ting Yan 提交于
Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 Jens Elofsson 提交于
* Update the ethosu-core-driver version. * Specify Arm Ethos-U version. The new version of ethos-u-core-driver requires the Ethos-U version to be specified. Add ETHOSU_ARCH as a variable to ethosu.inc and specify "u55" as the version for the Arm Corstone-300 target. * Update the ethos-u kernel to match new driver. Update the ethosu.cc with a call to ethosu_invoke_v3. Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
- 08 2月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 05 2月, 2022 1 次提交
-
-
由 TFLM-bot 提交于
-
- 04 2月, 2022 3 次提交
-
-
由 deqiangc 提交于
Temporarily disable the global default enforcement of pairing of AllocateTempXxx/DeallocateTemp (#908) * Temporarily disable the global enforcement of pairing of AllocateTemp/DeallocateTempXxx Temporarily disable the global enforcement of pairing of AllocateTemp/DeallocateTempXxx because there are some internal test that is using KernelRunner. BUG=https://b/217818824 * Disable default global enforcing of pair of AllocateTempXxx/DeallocateTemp * Revert xtensa svdf change
-
由 deqiangc 提交于
BUG=https://issuetracker.google.com/211673155Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 deqiangc 提交于
* Clean up GetInput/GetOutput api usage in squeeze, transpose, strided_slice op Clean up api usage in squeeze, transpose, strided_slice op by replacing GetInput/GetOutput with AllocateTempXXX/DeallocateTempXXX in micro context api in squeeze, transpose, strided_slice op Additionally, squeeze and transpose is using TfLiteTensor in Invoke, which is againts the recommendation and unnecessary. Replace them with the usage of TfLiteEvalTensor. This is a continuation of pr #894 to improve both readability and safety of getting tensors. The next step is to clean the GetInput/GetOutput in test_helpers.cc before enforcing the pair of AllocateTempXXX and DeallocateTemp globally. BUG=https://b/209453859 * Fix bluepill test: need to explictly use int32_t so that the right templated typeToTfLiteType is chosen and return the right type * Remove redundant include * Clean GetInput/GetOutput api usage in test_helpers * Enforce the pairing of AllocateTemp/DeallocateTemp for all kernel tests * Fix svdf/full_connected in cmsis_nn variant and conv in xtensa variant * Fix xtensa * Fix depthwise_conv in xtensa * Fix xtensa again * Fix compilation. Xtensa local test pass * Fix xtensa/full_connected not showing up locally * update cmsis_nn svdf * Fix pooling for hifi5 * hif5 errors seen only on server * Fix hifi5 again
-
- 03 2月, 2022 1 次提交
-
-
由 rockyrhodes 提交于
-
- 02 2月, 2022 6 次提交
-
-
由 Shlomi Regev 提交于
* Reshape op - replace slow byte-wise copy with memcpy() memcpy() usually comes optimized for a DSP/Micro and offers superior performance. BUG=217301557 * Add newline * Now add the newline at the right place. * Add #include <cstring> for memcpy() Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 Shlomi Regev 提交于
CXXFLAGS BUG=217301557
-
由 rockyrhodes 提交于
-
由 David Davis 提交于
* Add files from TfLite Added batch_matmul.h and tensor_utils_common.h that are needed for the BATCH_MATMUL operator. * Convert Colab hello_world training to local Python script. bug=#892 Use tensorflow 2.7.0 Add assertion checks. Add loss and MAE limits. Add pip install of modules. Add shell script to compare models. * Fix formatting errors not caught by VSCode. Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-
由 Ting Yan 提交于
-
由 deqiangc 提交于
Replace (almost) all kernel's GetInput/GetOutput with AllocateTempXXx and DeallocateTempXxx micro context api (#894) * Replace GetInput/GetOutput with micro context api * Replace GetInput/GetOutput in kernels with micro_context api. Ther are three kernels (strided_slice, squeeze, transpose)that are not changed yet because: they put the returned value of GetInput/GetOutput in an OpData structure that is persistent. It is better to have separate change to opdata of those kernels * Fix corner case in pad * Fix cmsis-nn * Fix xtensa, cmsis_nn, hexagon
-
- 01 2月, 2022 2 次提交
-
-
由 deqiangc 提交于
When upgrading the docker container, we see new format fix/check on existing python file. That is because we did not fix yapf version. This PR sets the yapf version to the latest 0.32.0, which is what the latest container is using BUG=https://issuetracker.google.com/216670232
-
由 TFLM-bot 提交于
-
- 28 1月, 2022 4 次提交
-
-
由 Nat Jeffries 提交于
* Create pad integration tests for SEANet. Modify generation script to include PAD operator and fix issue where shapes were being modified. Size > 2 is still valid since all weights for conv and transpose_conv have shapes with length 4. * Revert format changes removing newlines in generat_per_layer_tests.py
-
由 Ting Yan 提交于
* Add unit tests for 16-bit tanh and logistic kernels * Disable int16 logistic test for xtensa Xtensa optimized logistic kernel does not support int16 yet. Disable the test for it for now. * Use reference int16 logistic for xtensa and update test accordingly * Document source of test vectors and expected results
-
由 jwithers 提交于
* adding pandas and matplotlib * updating container tag to latest * fixing formatting issues that have been slipping through older ci setup Co-authored-by: Ndeqiangc <86809673+deqiangc@users.noreply.github.com>
-
由 deqiangc 提交于
Add AllocateTempInputTensor to replace GetInput() and Add AllocateTempOutputTensor to replace GetOutput() (#885) This corresonds to internal cl/424648139 Add more informative api AllocateTempInputTensor to replace GetInput() and AllocateTempOutputTensor to replace GetOutput(). These two new API call out the fact that what GetInput and GetOutput returns are in temporary buffer and need to call DeallocateTempTensor() afterwards. IF operator is the first OP to use these APIs. BUG=https://b/209453859Co-authored-by: Nmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
-