metadata
license: apache-2.0
tags:
- int8
- Intel® Neural Compressor
- PostTrainingStatic
datasets:
- squad
metrics:
- f1
INT8 DistilBERT base uncased finetuned on Squad
Post-training static quantization
This is an INT8 PyTorch model quantized with Intel® Neural Compressor.
The original fp32 model comes from the fine-tuned model distilbert-base-uncased-distilled-squad.
The calibration dataloader is the train dataloader. The default calibration sampling size 300 isn't divisible exactly by batch size 8, so the real sampling size is 304.
The linear module distilbert.transformer.layer.1.ffn.lin2 falls back to fp32 to meet the 1% relative accuracy loss.
Test result
| INT8 | FP32 | |
|---|---|---|
| Accuracy (eval-f1) | 86.1069 | 86.8374 |
| Model size (MB) | 74.7 | 265 |
Load with Intel® Neural Compressor:
from neural_compressor.utils.load_huggingface import OptimizedModel
int8_model = OptimizedModel.from_pretrained(
'Intel/distilbert-base-uncased-distilled-squad-int8-static',
)