Anton Lokhmotov
Anton Lokhmotov
In the [official results dashboard](https://mlcommons.org/en/training-normal-11/), all links to code and systems for Habana Labs lead to https://github.com/mlcommons/training_results_v1.1/blob/master/HabanaLabs/ (broken), while they should be leading to https://github.com/mlcommons/training_results_v1.1/blob/master/Intel-HabanaLabs/ /cc @bitfort
The [MLPerf Results Messaging Guidelines](https://github.com/mlcommons/policies/blob/master/MLPerf_Results_Messaging_Guidelines.adoc#7-timing-for-results-disclosures) contain the following: > Submitters (i.e., those who have submitted results for review and verification by MLCommons) are not allowed to publish any results for...
In the MLPerf Inference v2.1 round, Qualcomm and their partners submitted a number of RetinaNet results, which the Review Committee eventually accepted to the Open division under the Preview category....
The [results table](https://mlperf.org/inference-results/) has a Notes column. However, at the moment it is filled in manually. For populating this column automatically, I propose allowing to create a Tweet-sized file called...
At the moment, we place `user.conf` under `measurements/` e.g. `closed/Intel/measurements/clx_9282-2s_openvino-linux/ssd-small/Server`. It can be argued, however, that it belongs under `results` e.g. `closed/Intel/results/clx_9282-2s_openvino-linux/ssd-small/Server`. I don't think we should strictly require all...
As [agreed](https://docs.google.com/presentation/d/1urn23orVQq5Z5Jz8ZJf_o1dASlhIWDSLYi2fyuEY4hw/edit#slide=id.g915930d977_0_0) at Submitters' WG meeting on 1/Sep/2020: > WG: compliance tests for open submission by review committee request only
Add support for [Cedric Nugteren](http://www.cedricnugteren.nl)'s [CLBlast](https://github.com/CNugteren/CLBlast) library.
Add support for the popular open-source [clBLAS](https://github.com/clMathLibraries/clBLAS) library. Thanks to [Cedric Nugteren](http://www.cedricnugteren.nl/) for the suggestion.
GEMMbench currently uses the `clCreateCommandQueue()` API function, which is [deprecated in OpenCL 2.0](https://www.khronos.org/registry/cl/sdk/2.0/docs/man/xhtml/deprecated.html). Add support for `clCreateCommandQueueWithProperties()` in OpenCL 2.0 (perhaps using macros as suggested [here](http://stackoverflow.com/questions/28500496/opencl-function-found-deprecated-by-visual-studio)).
A companion to https://github.com/mlcommons/inference/issues/1581 A workaround is provided with two PRs: - https://github.com/mlcommons/power-dev/pull/325 - https://github.com/mlcommons/inference/pull/1584 This will work for v4.0 submitters who have already generated some experiments. Ideally, we should...