Anton Lokhmotov
Anton Lokhmotov
We should investigate how to enable fp16 e.g. in benchmarks. This [TensorFlow issue](https://github.com/tensorflow/benchmarks/issues/77) may provide some clues.
When running TF benchmarks on an Intel Xeon, warnings appear: ``` W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and...
On one of my platforms, OpenCL programs only run under super user e.g. ``` # ck run program:arm-mali-sgemm ``` It used to be the case that OpenCL programs could be...
We've been using the standard [`pycocotools`](https://pypi.org/project/pycocotools/) Python package for calculating the Object Detection accuracy since MLPerf Inference v0.5. It used to be OK for SSD-ResNet34 and SSD-MobileNet-v1, but it is...
TensorRT 3.0 only works with CUDA 9.0. However, I currently have both CUDA 9.0 and CUDA 9.1 installed: ``` $ ck show env --tags=cuda Env UID: Target OS: Bits: Name:...
At the moment, the widget can only visualise already obtained data. After #12, we should implement a mode similar to our [desktop demo](https://github.com/dividiti/ck-crowdsource-dnn-optimization):  This mode would list all available...
The versions of OpenCV we currently support are quite old: ``` $ ck search package --tags=lib,opencv,channel-stable ck-math:package:lib-opencv-3.4.5 ck-math:package:lib-opencv-3.3.0 ck-math:package:lib-opencv-2.4.13.2 ck-math:package:lib-opencv-3.4.3 ck-math:package:lib-opencv-3.2.0 ``` The [latest releases](https://opencv.org/releases/) are 3.4.10 and 4.3.0. (Intel...
I have noticed that when Boost is installed as a Caffe dependency (and possibly standalone), in addition to `compiled-by-gcc` and and `compiled-by-gcc-
Probably need to set `OPENCL_LIBRARIES`. ``` $ ck install package:lib-clblas-master-universal ... -- Boost version: 1.64.0 -- Found the following Boost libraries: -- program_options -- Boost_PROGRAM_OPTIONS_LIBRARY: /home/anton/CK_TOOLS/lib-boost-1.64.0-gcc-6.3.0-linux-64/install/lib/libboost_program_options.a -- CMAKE_C_COMPILER_MACHINE: aarch64-linux-gnu --...
On Odroid-XU3 with 4 "big" Cortex-A15 cores and 4 "LITTLE" Cortex-A7 cores, OpenBLAS configures itself with the maximum number of threads equal to the total number of cores, i.e. 8...