- 07 8月, 2021 3 次提交
-
-
由 deqiangc 提交于
* Add a new test case for conv operator (#329) This new test case is based on issue #329 to increase coverage on optimized kernel's data precision for conv operator. In this test, input, output and filter are all 8 bits and filter tensor is of dimension 8x3x3x3 with different scales per output channel. TESTED= local test with x86 and HiFi4. * Move large test data variable into its own folder and use one flat top namespace instead of nested namespace. BUG=195779890
-
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. * Arduino library creation for CI build. Issue# 256 * fix bug with include file path flattening * fix examples path generation Some files in examples source and headers list are not in the examples directory. They can be tensorflow files or third-party files. The code is corrected so that these files end up inside the generated directory tree. Previously, these files were being placed outside the generated directory tree and were thus invisible to follow-on targeted generation scripts. * remove/revert files that go into the arduino-examples repo * Format with: ``` tensorflow/lite/micro/tools/ci_build/test_code_style.sh --fix_formatting ``` * request changes to PR * Removed symbolic link for micro_speech example from Makefile. Added additional include flags to Makefile. Adjusted header file transformations during base tree creation. * Fix formatting. * Move Arduino specific code to seperate repo Remove ARDUINO defines. Fix incorrect NDEBUG use. Issue #256 * Fix cortex_m_generic makefile Fix typo in cortex_m_generic makefile so that the correct list of header files is generated by the create_tflm_tree.py script. bug=Arduino library creation external to the tflite-micro repository #256 Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 Artem Tsvetkov 提交于
-
- 06 8月, 2021 3 次提交
-
-
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. * Arduino library creation for CI build. Issue# 256 * fix bug with include file path flattening * fix examples path generation Some files in examples source and headers list are not in the examples directory. They can be tensorflow files or third-party files. The code is corrected so that these files end up inside the generated directory tree. Previously, these files were being placed outside the generated directory tree and were thus invisible to follow-on targeted generation scripts. * remove/revert files that go into the arduino-examples repo * Format with: ``` tensorflow/lite/micro/tools/ci_build/test_code_style.sh --fix_formatting ``` * request changes to PR * Removed symbolic link for micro_speech example from Makefile. Added additional include flags to Makefile. Adjusted header file transformations during base tree creation. * Fix formatting. * Move Arduino specific code to seperate repo Remove ARDUINO defines. Fix incorrect NDEBUG use. Issue #256 Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 TFLM-bot 提交于
* Sync from upstream TF. * Fix build errors. * more fixes. * one more fix. Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 Nat Jeffries 提交于
-
- 05 8月, 2021 1 次提交
-
-
由 Nat Jeffries 提交于
* Fix typo in test_code_style.sh * Format python code. * Reset code style.
-
- 03 8月, 2021 3 次提交
-
-
由 Artem Tsvetkov 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. * Arduino library creation for CI build. Issue# 256 * fix bug with include file path flattening * fix examples path generation Some files in examples source and headers list are not in the examples directory. They can be tensorflow files or third-party files. The code is corrected so that these files end up inside the generated directory tree. Previously, these files were being placed outside the generated directory tree and were thus invisible to follow-on targeted generation scripts. * remove/revert files that go into the arduino-examples repo * Format with: ``` tensorflow/lite/micro/tools/ci_build/test_code_style.sh --fix_formatting ``` * request changes to PR * Removed symbolic link for micro_speech example from Makefile. Added additional include flags to Makefile. Adjusted header file transformations during base tree creation. Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 Lauren Murphy 提交于
On commit 5c7c11e of the tensorflow repo at tensorflow/lite/micro/tools/make/Makefile:664, MICROLITE_CC_SRCS contains MICROLITE_CC_KERNEL_SRCS. In the new tflite-micro repo, it does not contain MICROLITE_CC_KERNEL_SRCS. This commit adds MICROLITE_CC_KERNEL_SRCS to tensorflow/lite/micro/examples/hello_world/zephyr_riscv/Makefile.inc so that the generated Zephyr CMakeLists.txt doesn't miss these files. Similar fix made to corresponding file for magic_wand sample. Signed-off-by: NLauren Murphy <lauren.murphy@intel.com> Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
- 28 7月, 2021 6 次提交
-
-
由 Artem Tsvetkov 提交于
-
由 cad-audio 提交于
* REF_CODE_REFACTOR: hard_swish Refactoring the reference code for hard_swish operator. BUG=refactoring existing code. * Remove uint8 support. * fix formatting. Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 cad-audio 提交于
* REF_CODE_REFACTOR: logicalAND and logicalOR Refactoring the reference code for logicalAND and logicalOR operators. BUG=refactoring existing code. * Fix formatting. * fix bad merge (copyright year). Co-authored-by: NAdvait Jain <advaitjain@google.com> Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
由 Pete Warden 提交于
* Fixes for RISCV build * Consolidate compiler flags * Added RISC-V continuous integration script Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
由 cad-audio 提交于
* REF_CODE_REFACTOR: logistic Refactoring the reference code for logistic operator. BUG=refactoring existing code. * Fix formatting Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 cad-audio 提交于
Updated the script to download HiFi 4 Neural Network Library v2.4.1. Removed the hifi4 patch Updates in v2.4.1 library: + Fixed the issue related to special case handling in xa_nn_dot_prod_16x16_asym8s. Moved the special cases handling for vec_length=8/32 to the start of the function. + Removed legacy condition checks for depthwise_conv kernels. + Fixed the non-bit-exact issue related to xa_nn_conv2d_std_per_chan_sym8sxasym8s. updated to select correct vec values in case of vec_align_val is 1 and
-
- 27 7月, 2021 3 次提交
-
-
由 Artem Tsvetkov 提交于
-
由 cad-audio 提交于
* REF_CODE_REFACTOR: activations.cc Refactoring the reference code for activations operators. BUG=refactoring existing code. * Remove uint8 support. * Ran `test_code_style.sh --fix_formatting` * Fix the build. Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 Pete Warden 提交于
* Fixes for RISCV build * Consolidate compiler flags
-
- 24 7月, 2021 2 次提交
-
-
由 Advait Jain 提交于
With http://cl/386537127 (tensorflow/tensorflow@d4987a9) we do not run into the type-limits warning and -Wno-type-limits can be removed. BUG=http://b/194523886
-
由 Michael O'Cleirigh 提交于
Fixes #294 Copy from xtensa ci scripts to run the test script inside of a docker container. We use the espressif/idf:release-v4.3 image. Add a build that builds the full tensorflow-microlite.a library. The person detection example is more complicated and not fixed here. Co-authored-by: NMichael O'Cleirigh <michael.ocleirigh@gmail.com> Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
- 23 7月, 2021 4 次提交
-
-
由 Patrik Laurell 提交于
* Port int16 support to MEAN kernel Change-Id: I7633099ea261ccdca424b01a591738f6dfae62eb * Port tests for int16 MEAN kernel Change-Id: I2731476dccb33d2a429e9a9e2b7b86efb3a71e16 * Add test for int16 mean, different input and output scale and zero point * Fix bug in calculating multiplier and shift for int16 case Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-
由 Advait Jain 提交于
* Groundwork to enable python format checks (via yapf) as part of the CI. With this change: * we are set up to use yapf as part of the docker container * pigweed patch is updated to use the google Python style (pep8 with indent of 2 spaces) * some documentation updates. The code still needs to be formatted, and then the Python formatting check will be turned on as part of ci. BUG=http://b/194404216 * Format all the code. Once this change is merged, we will turn on Python formatting check as part of the CI. * minor formatting update. * Turn on Python formatting checks as part of CI. Also, add an option to fix the formatting via the test_code_style.sh script. BUG=http://b/194404984 and http://b/194404216
-
由 Advait Jain 提交于
* Groundwork to enable python format checks (via yapf) as part of the CI. With this change: * we are set up to use yapf as part of the docker container * pigweed patch is updated to use the google Python style (pep8 with indent of 2 spaces) * some documentation updates. The code still needs to be formatted, and then the Python formatting check will be turned on as part of ci. BUG=http://b/194404216 * Format all the code. Once this change is merged, we will turn on Python formatting check as part of the CI. * minor formatting update.
-
由 Advait Jain 提交于
With this change: * we are set up to use yapf as part of the docker container * pigweed patch is updated to use the google Python style (pep8 with indent of 2 spaces) * some documentation updates. The code still needs to be formatted, and then the Python formatting check will be turned on as part of ci. BUG=http://b/194404216
-
- 21 7月, 2021 3 次提交
-
-
由 Advait Jain 提交于
* Remove distinction between fusion_f1 and hifi4. Hifi4, Fusion F1 and Hifi3Z all use the same underlying xannlib hifi4 optimizations and with this change we can specify TARGET_ARCH=hifi4 for each of these cores. BUG=http://b/194225949 * Fix the build
-
由 Kun Lu 提交于
Use base class version of AllocateFromTail() instead of derived class version in FlatBufferVectorToTfLiteTypeArray() to solve the issue Signed-off-by: NKun-Lu <kun.lu@ibm.com>
-
由 Artem Tsvetkov 提交于
-
- 17 7月, 2021 2 次提交
-
-
由 cad-audio 提交于
* Cadence HiFi 5 Neural Network Library v1.6.0 updated the download script to use hifi5 v1.6.0 archive. depthwise_conv kernel is updated and patch is not required now. quantize optimized kernel is renamed in the v1.6.0, updated with new kernel name. * Fix formatting. Co-authored-by: Nbhanu prakash bandaru venkata <bhanup@cadence.com> Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
由 Artem Tsvetkov 提交于
* Updated utility functions to work with embARC MLI Library 2.0 * Updated copyrights in several files. * Minor fix for mli_tf_utils.h * Update mli_tf_utils.h
-
- 16 7月, 2021 3 次提交
-
-
由 Artem Tsvetkov 提交于
* Updated pooling kernel to work with embARC MLI Library 2.0 * Update pooling.cc copyright.
-
由 Artem Tsvetkov 提交于
* Updated fully_connected kernel to work with embARC MLI Library 2.0 * Update fully_connected.cc * Minor fix for fully_connected activation functions.
-
由 Artem Tsvetkov 提交于
* Updated depthwise_conv kernel to work with embARC MLI Library 2.0 * Update depthwise_conv.cc * Minor fix for depthwise conv activation functions.
-
- 15 7月, 2021 3 次提交
-
-
由 Artem Tsvetkov 提交于
* Updated convolution kernel to work with embARC MLI Library 2.0 * Update conv.cc * Minor fix for conv2d activation functions.
-
由 Advait Jain 提交于
BUG=http://b/193437031
-
由 Advait Jain 提交于
* Have a TFLM-specific version of lite/kernels/op_macros.h * After this change, we can remove the TFLM-specific code from upstream Tensorflow. * Both the TfLite and TFLM implementations of op_macros.h will be simplified. BUG=http://b/187728891 NO_CHECK_TFLITE_FILE=lite/kernels/kernel_util.cc will be modified in upstream as well * Fix target-specific makefile build.
-
- 14 7月, 2021 2 次提交
-
-
由 Advait Jain 提交于
The corresponding internal change is http://cl/384575105 BUG=http://b/188064725
-
由 cad-audio 提交于
* XTENSA_CODE_REFACTOR: depthwise_conv Refactored the xtensa hifi optimized code for depthwise_conv operator. * Small fixes for the CI. Co-authored-by: Nbhanu prakash bandaru venkata <bhanup@cadence.com> Co-authored-by: NAdvait Jain <advaitjain@google.com>
-
- 13 7月, 2021 2 次提交
-
-
由 Advait Jain 提交于
* Force FLATBUFFERS_LOCALE_INDEPENDENT=0 With this change, we no longer need a #define in the sources that directly or indrectly include flatbuffers/base.h BUG=http://b/193264978 * Fix xtensa script.
-
由 Alan Green 提交于
* micro_profiler.cc: add CSV output Adds a method to log output in CSV form, suitable for input to a spreadsheet. * micro_profiler.cc: remove ms from CSV output Removes milliseconds from CSV output. Co-authored-by: NAdvait Jain <advaitjain@users.noreply.github.com>
-