Don't hesitate to get in touch if you encounter any issues or would like to discuss this community project!
Please report if this CK component works: 1▲
or fails: 0▲
Sign up to be notified when artifacts are shared or updated!
Live CK scoreboards are connected with CK solutions
and adaptive containers to make it easier to participate in collaborative AI/ML/SW/HW benchmarking.
See the CK white paper for more details.
Benchmark results (performance): P_*_SS - Single Stream in milliseconds, P_*_MS - MultiStream in no. streams, P_*_S - Server in QPS, P_*_O - Offline in inputs/second. Benchmark results (accuracy): A_IC* - Top-1, A_OD* - mAP, A_NMT* - BLEU. cK components: packages, software detection plugins. Image Classification: IC1 - ImageNet, MobileNet-v1, IC2 - ImageNet, ResNet-50 v1.5. Object detection: OD1 - COCO, SSD w/ MobileNet-v1, OD2 - COCO 1200x1200, SSD w/ ResNet-34. Translation: NMT - WMT E-G, NMT. Form Factor: FF_M - Mobile/Handheld, FF_D - Desktop/Workstation, FF_S - Server, FF_E - Edge/Embedded. These are not official results but a snapshot to collaboratively reproduce results and add portable workflows! MLPerf name and logo are trademarks. See www.mlperf.org for more information.