diff options
author | 김수진/On-Device Lab(SR)/Engineer/삼성전자 <sjsujin.kim@samsung.com> | 2019-02-19 10:11:39 +0900 |
---|---|---|
committer | 오형석/On-Device Lab(SR)/Staff Engineer/삼성전자 <hseok82.oh@samsung.com> | 2019-02-19 10:11:39 +0900 |
commit | 703db79b8efc5be3d9d1517e7b0a04806366c07a (patch) | |
tree | 8418fd1ce0e8a131d16199b85ae8ad0db21cc70c /tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu | |
parent | f45f9d68fd0c394b6a471ad9c68698909fd86253 (diff) | |
download | nnfw-703db79b8efc5be3d9d1517e7b0a04806366c07a.tar.gz nnfw-703db79b8efc5be3d9d1517e7b0a04806366c07a.tar.bz2 nnfw-703db79b8efc5be3d9d1517e7b0a04806366c07a.zip |
[neurun] Use legacy tflite method in ConvolutionLayer for cpu (#4416)
As we've updated the tensorflow to v1.12, `ConvolutionLayer` kernel for `cpu` was updated.
However,`ConvolutionLayer` of cpu kernel for `quant8` makes wrong matching results when running genereated tests.
Legacy methods of that work well for `quant8`.
`Android NN` master barnch Conv2D also keep using legacy methods.
For consistency and being useful to support `quant8`, we can use legacy methods instead of newly added methods.
(Added TODO comment about changing to new version)
Signed-off-by: sjsujinkim <sjsujin.kim@samsung.com>
Diffstat (limited to 'tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu')
-rw-r--r-- | tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu | 9 |
1 files changed, 0 insertions, 9 deletions
diff --git a/tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu b/tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu index e229902e3..a6ce26c35 100644 --- a/tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu +++ b/tests/nnapi/nnapi_gtest.skip.armv7l-linux.neurun.cpu @@ -22,14 +22,6 @@ ValidationTestExecution.StartCompute ValidationTestExecution.EventWait GeneratedTests.add* GeneratedTests.argmax* -GeneratedTests.conv_quant8_channels -GeneratedTests.conv_quant8_channels_weights_as_inputs -GeneratedTests.conv_quant8_large -GeneratedTests.conv_quant8_large_weights_as_inputs -GeneratedTests.conv_quant8 -GeneratedTests.conv_quant8_overflow -GeneratedTests.conv_quant8_overflow_weights_as_inputs -GeneratedTests.conv_quant8_weights_as_inputs GeneratedTests.depth_to_space* GeneratedTests.depthwise_conv2d* GeneratedTests.depthwise_conv @@ -82,7 +74,6 @@ GeneratedTests.tensorflowmax_ex* GeneratedTests.reduce_sum_ex* GeneratedTests.topk_v2* # Unexpected result -GeneratedTests.conv_quant8_2 GeneratedTests.split* GeneratedTests.transpose_conv* GeneratedTests.pack* |