ctuning/cbench-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows repository overview

This is a demo of a portable benchmarking pipeline to automate and simplify the MLPerf benchmark⁠ across diverse platforms provided by volunteers similar to SETI@home: https://cKnowledge.io/solution/demo-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows⁠

You can run at as follows:

docker run ctuning/cbench-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows /bin/bash -c "cb benchmark demo-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows"

You can then check the official results (black dots) and crowd-results from volunteers (red dots) at the public dashboard: https://cKnowledge.io/result/sota-mlperf-object-detection-v0.5-crowd-benchmarking⁠

You can also run this benchmark in the interactive mode via the CK GUI. Run the following command:

docker run -p 4444:4444 -it ctuning/cbench-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows /bin/bash -c "cr start -h 0.0.0.0"

Then navigate to this this GUI page⁠ and click the "Auto-benchmark" button.

Finally, you can test this solution interactively as follows:

sudo docker run -it ctuning/cbench-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows /bin/bash

$ cb benchmark demo-obj-detection-mlperf-coco-tf-cpu-benchmark-linux-portable-workflows

If you encounter any issues or have suggestions for improvement, please do not hesitate to get in touch⁠ to help our community project!