PII_Bert / real_test_inference_metrics_T4_onnx.csv
PuxAI's picture
Add real Test Dataset inference benchmarks (ONNX Runtime)
cc58d74 verified
model,engine,provider,gpu_used,test_samples,total_valid_tokens,total_time_seconds,latency_ms_per_sample,throughput_samples_per_sec,throughput_tokens_per_sec
google-bert-bert-base-multilingual-cased-onnx,onnxruntime,CUDAExecutionProvider,Tesla T4,4677,207365,133.99,28.65,34.91,1547.66