Latency Speedup on SQuAD v1.1 (Question Answering)
1.34Latency SpeedupFast Post-Training Pruning Framework
Evaluation results
| Method | Links | |
|---|---|---|
| Fast Post-Training Pruning FrameworkBatch size=256, Backbone=BERT_BASE, Hardware=NVIDIA V100 GPU, Framework=PyTorch, Accuracy degradation constraint=at most 1%2022.03 | 1.34 | |
| Fast Post-Training Pruning FrameworkBatch size=32, Backbone=BERT_BASE, Hardware=NVIDIA V100 GPU, Framework=PyTorch, Accuracy degradation constraint=at most 1%2022.03 | 1.33 |