PuxAI commited on
Commit
5a9e593
·
verified ·
1 Parent(s): fc820c1

Add real Test Dataset inference benchmarks from T4 GPU

Browse files
Files changed (1) hide show
  1. real_test_inference_metrics_T4.csv +7 -0
real_test_inference_metrics_T4.csv ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ model,gpu_used,test_samples,total_valid_tokens,total_time_seconds,latency_ms_per_sample,throughput_samples_per_sec,throughput_tokens_per_sec
2
+ bert-base-cased,Tesla T4,2176,111284,59.43,27.31,36.62,1872.65
3
+ roberta-base,Tesla T4,2176,102023,57.15,26.26,38.08,1785.32
4
+ distilbert-base-cased,Tesla T4,2176,111284,30.46,14.0,71.43,3652.97
5
+ albert-base-v2,Tesla T4,2176,103998,70.5,32.4,30.87,1475.21
6
+ FacebookAI-xlm-roberta-base,Tesla T4,2176,106333,57.45,26.4,37.87,1850.75
7
+ google-bert-bert-base-multilingual-cased,Tesla T4,2176,111171,61.12,28.09,35.6,1818.9