- 24 3月, 2020 6 次提交
-
-
由 cc 提交于
-
由 xiebaiyuan 提交于
-
由 huzhiqiang 提交于
-
由 zhupengyang 提交于
-
由 xiebaiyuan 提交于
* [LITE][OPENCL][Image]optimise conv2d 5x5 7x7,test=develop * [LITE][OPENCL][Image]optimise conv2d 5x5 7x7,test=develop * [LITE][OPENCL][Image]optimise conv2d 5x5 7x7,test=develop
-
由 Yuan Shuai 提交于
* [LITE][OPENCL] comment vlog for opencl kernel. test=develop
-
- 23 3月, 2020 5 次提交
-
-
由 HappyAngel 提交于
* fxi ff, test=develop * delete bgr_flip, test=develop
-
由 xiaogang 提交于
-
由 Yuan Shuai 提交于
* Ehance mobilenetv1/v2, add std_dev computation. test=develop * Ehance cxx mobile_light demo, add std_dev computation. test=develop
-
由 mapingshuo 提交于
* rm grad code
-
由 zhupengyang 提交于
-
- 22 3月, 2020 1 次提交
-
-
由 Yuan Shuai 提交于
* [LITE][OPENCL] clean code for opencl. test=develop * [LITE][PROFILER] Enhance Precision Profiler. test=develop * delete useless var in profiler. test=develop * add ocl header. test=develop
-
- 20 3月, 2020 5 次提交
-
-
由 hong19860320 提交于
-
由 cc 提交于
-
由 huzhiqiang 提交于
-
由 cc 提交于
-
由 mapingshuo 提交于
* rm grad code * add mul_grad, test=develop
-
- 19 3月, 2020 3 次提交
-
-
由 huzhiqiang 提交于
* add strip lib in ios * add strip lib in ios test=develop
-
由 HappyAngel 提交于
* fix memory leaky, test=develop
-
由 Yuan Shuai 提交于
* [LITE][OPENCL] Change fp32 fc to fp16's. test=develop * fix act in conv3x3opt opencl kernel. test=develop * [LITE][CMAKE] fix cmake for opencl tiny_pubish compile. test=develop
-
- 18 3月, 2020 7 次提交
-
-
由 Yuan Shuai 提交于
* Fix OpenCL kernel of exp, tanh, * Fix layout pass for opencl backend, * Add how to debug cl kernel. test=develop.
-
由 huzhiqiang 提交于
-
由 cc 提交于
-
由 zhupengyang 提交于
-
由 HappyAngel 提交于
-
由 huzhiqiang 提交于
-
由 zhupengyang 提交于
This reverts commit 670fcc3c.
-
- 17 3月, 2020 9 次提交
-
-
由 zhupengyang 提交于
-
由 Wilber 提交于
add cuda c++ doc
-
由 HappyAngel 提交于
*fix yuv resize bug, test=develop * fix format, test=develop * fix build error, test=develop * fix resize, test=develop * move anakin rst, test=develop
-
由 huzhiqiang 提交于
-
由 Wilber 提交于
- 增加cuda c++ demo. - 考虑到检测模型尾部一般是multiclass_nms,该kernel为host,如果fetch kernel为cuda的话,则会在此处插入无用的io_copy(host->cuda),由于该原因,注释掉fetch的cuda kernel. 默认使用host的fetch kernel. 此处暗中进行的行为:每次predictor run完,都会默认把数据从cuda拷贝到cpu
-
由 Santa An 提交于
* * support download bm_sdk, test=develop * [LITE][BM] add slice op * [LITE][BM] fix concat issue * [LITE][BM] support reduce full ops, test=develop * [LITE][BM] change test_resnet50 to change test_classify * [LITE][BM] add cast op * [LITE][BM] add reduce and other ops, test=develop * [LITE][BM] add reduce,cast and other ops, test=develop
-
由 石晓伟 提交于
* feature: buffer sharing, test=develop * fix the warning of buffer.h, test=develop * test cuda only if necessary, test=develop
-
由 Wilber 提交于
add cuda ci. Organize cuda compilation products.
-
由 mapingshuo 提交于
* add sgd op, test=develop * test=develop
-
- 16 3月, 2020 4 次提交
-
-
由 Yuan Shuai 提交于
* [LITE][OPENCL] Change fp32 fc to fp16's. test=develop * fix act in conv3x3opt opencl kernel. test=develop * [LITE][ARM] comment assembly of layout armv7 cpu. test=develop
-
由 hong19860320 提交于
-
由 mapingshuo 提交于
* add backend * add kernel * add grad kernel
-
由 cc 提交于
-