- 07 1月, 2020 1 次提交
-
-
由 huzhiqiang 提交于
-
- 06 1月, 2020 2 次提交
-
-
由 石晓伟 提交于
-
由 liu zhengxi 提交于
* alter the api name from cpu to x86, test=develop * correct the step_rnn model test, test=develop
-
- 03 1月, 2020 4 次提交
-
-
由 xiebaiyuan 提交于
* [mobile][opencl] suite model male2fe ,support a type element_mul ,test=mobile * [mobile][opencl] add test4 male2fe ,test=mobile
-
由 zhupengyang 提交于
test=develop
-
由 Wilber 提交于
temporarily remove cuda fc fuse because we don't support cuda fc now
-
由 hong19860320 提交于
-
- 02 1月, 2020 3 次提交
-
-
由 石晓伟 提交于
-
由 GaoWei8 提交于
* Enhance fc_fuse_pass to enable fusing relu to fc_op test=develop * restrict fusing relu in x86 test=develop
-
由 hong19860320 提交于
-
- 31 12月, 2019 5 次提交
-
-
由 xiebaiyuan 提交于
-
由 NazgulLee 提交于
-
由 Wilber 提交于
X86 and cuda compile simutaneously cmake .. -DCMAKE_BUILD_TYPE=RelWithDebInfo -DWITH_MKL=ON -DLITE_WITH_CUDA=ON -DWITH_MKLDNN=OFF -DLITE_WITH_X86=ON -DLITE_WITH_PROFILE=OFF -DWITH_LITE=OFF -DLITE_WITH_LIGHT_WEIGHT_FRAMEWORK=OFF -DWITH_PYTHON=OFF -DWITH_TESTING=ON -DLITE_WITH_ARM=OFF -DLITE_ON_TINY_PUBLISH=OFF -DCUDNN_ROOT=/usr/local/cudnn/ -DLITE_BUILD_EXTRA=ON (#2708) x86 and cuda compile simutaneously
-
由 zhupengyang 提交于
test=develop
-
由 hong19860320 提交于
* Fix the compiling error which occurs when specify the ddk_root path and build for huawei NPU. * Refine the registration of op bridges and make it similar to the registration of op and kernel. * Refine the interfaces of the graph and node for op bridges, and support creating constant and data node automatically according to the attribute 'persistable' of the target tensor. * Add the unit test of the scale and softmax op bridge for NPU.
-
- 30 12月, 2019 2 次提交
-
-
由 juncaipeng 提交于
* fix yolov3 bug when run several times, test=develop
-
由 Yiqun Liu 提交于
Optimize the execution of RuntimeProgram by saving the bool whether the op is feed/fetch op. (#2703) test=develop
-
- 29 12月, 2019 1 次提交
-
-
由 xiebaiyuan 提交于
-
- 28 12月, 2019 3 次提交
-
-
由 xiebaiyuan 提交于
[mobile][opencl] make model attr support issue more readable ,fastfail before cxxlib err ,test=mobile (#2697)
-
由 Jiaying Zhao 提交于
-
由 huzhiqiang 提交于
-
- 27 12月, 2019 6 次提交
-
-
由 xiebaiyuan 提交于
* [mobile][opencl]optimise log print , ues kNOLOG to close develop time logs ,test=mobile * [mobile][opencl]optimise log print , ues kNOLOG to close develop time logs ,test=mobile
-
由 石晓伟 提交于
-
由 yiicy 提交于
-
由 huzhiqiang 提交于
-
由 hong19860320 提交于
-
由 huzhiqiang 提交于
remove test_models ci projects, because these project hass been removed in ci test test=develop (#2669)
-
- 26 12月, 2019 6 次提交
-
-
由 xiebaiyuan 提交于
* [mobile][opencl]common deepwise conv,test=mobile * [mobile][opencl]revert deepwise 3x3 for stable ,test = mobile * [mobile][opencl]format convkernel.inc.cl with clang-format ,test = mobile * [mobile][opencl] suite 1*X Y element_y ,test=mobile * [mobile][opencl] add whole print method for cl_image ,test=mobile
-
由 Wilber 提交于
-fix fluid-lite-subgraph x86 compile error - Replace FLAGS with environment variables
-
由 Jiaying Zhao 提交于
-
由 Jiaying Zhao 提交于
[Mobile][ARM]Fix memory leaks when using input tensor created with external pointer in lod_mode. (#2681)
-
由 xiaogang 提交于
* feat: add multi_thread ut
-
由 zhupengyang 提交于
test=develop
-
- 25 12月, 2019 6 次提交
-
-
由 juncaipeng 提交于
add clear for tensor
-
由 juncaipeng 提交于
* fix op inputs and outputs type, test=develop
-
由 Wilber 提交于
optimize softmax cuda kernel
-
由 juncaipeng 提交于
-
由 hong19860320 提交于
-
由 Yiqun Liu 提交于
* Remove GEMM padding in fc_compute. test=develop * Write a common ParallelFor function to run the for loop in parallel. * Add the codes of padding GEMM back in fc. * Refine the code of fc when padding_weight is false to avoid the definition of temporary Tensor. * Refine the unit test of fc and add testing case of padding and parallel. test=develop * Enable more test cases in common fc unittest, including padding and parallel for x86 target. * Remove the fc test under kernels/x86. test=develop * Disable relu in test of fc for non-x86 target. test=develop * Change the eps of arm. test=develop
-
- 24 12月, 2019 1 次提交
-
-
由 zhupengyang 提交于
-