Anton Lokhmotov

That's a good idea. BTW, I think the "INFO" page is completely missing when running the dashboard widget locally? ``` $ ck pull repo:ck-web $ ck display dashboard ```

A much better approach to copying would be to create a new `ck-math:package:lib-opencv` with _variations_ for different versions and options. For example: - [`package:lib-openvino`](https://github.com/ctuning/ck-openvino/blob/master/package/lib-openvino/.cm/meta.json#L95) - [`package:lib-armnn`](https://github.com/ctuning/ck-armnn/blob/master/package/lib-armnn/.cm/meta.json#L160)

The same tag gets added when compiling Boost standalone: ``` $ ck install package:lib-boost-1.64.0-min-for-caffe $ ck show env --tags=lib,boost,v1.64 Env UID: Target OS: Bits: Name: Version: Tags: 822b4e5283a69e6f linux-64 64...

@twang15 It's a typo: should be "crow**d**tuning", not "crowtuning". Please try: ``` $ ck pull repo:ck-crowdtuning ``` Did you copy this from somewhere in the documentation? If so, we should...

Perhaps it's somewhat similar to https://github.com/mlperf/inference/issues/334?

By the way, the above mentioned postprocessing issue also needs to be fixed but here's a quick workaround (installing the LoadGen Python module to userspace): ``` $ python -m pip...

If a model requires special steps (e.g. SSD-MobileNet), then it should motivate us even more to generalise.

Conversion steps for SSD-based models are documented [here](https://github.com/ctuning/ck-mlperf/blob/master/package/model-tflite-mlperf-ssd-mobilenet/README.md).

I wonder if the same conversion steps will work for [SSD-ResNet](https://zenodo.org/record/3060467/files/ssd_resnet-34_from_onnx.zip).

This can be reproduced as follows. 0. Install [CK-Caffe](github.com/dividiti/ck-caffe). 1. Install the ViennaCL master package: ``` $ ck install package:lib-viennacl-master-src --target_os=android21-arm64 ... Environment entry added (c3153bc343548550)! Recording CK configuration to...