Fix gemm: fix workloads of batch/group gemm; add task_dir in perf_engine; multiply qps with 1000

bytedance / ByteMLPerf

AI Accelerator Benchmark focuses on evaluating AI Accelerators from a practical production perspective, including the ease of use and versatility of software and hardware.

https://bytemlperf.ai/

Apache License 2.0

188 stars 50 forks source link

Fix gemm: fix workloads of batch/group gemm; add task_dir in perf_engine; multiply qps with 1000 #91

Closed kevinsouthByteDance closed 3 weeks ago

kevinsouthByteDance commented 3 weeks ago

Fix the workloads of gemm, batch gemm, group gemm,
Add the task_dir in perf_engine to support other different task directory
Multiply qps with 1000

CLAassistant commented 3 weeks ago

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}