Intel/bert-base-uncased-finetuned-swag-int8-static-inc

INT8 bert-base-uncased-finetuned-swag

Post-training static quantization

This is an INT8 PyTorch model quantized with huggingface/optimum-intel through the usage of Intel® Neural Compressor. The original fp32 model comes from the fine-tuned model thyagosme/bert-base-uncased-finetuned-swag.

The calibration dataloader is the train dataloader. The default calibration sampling size 100 isn't divisible exactly by batch size 8, so the real sampling size is 104.

The linear modules bert.encoder.layer.2.output.dense, bert.encoder.layer.5.intermediate.dense, bert.encoder.layer.9.output.dense, bert.encoder.layer.10.output.dense fall back to fp32 to meet the 1% relative accuracy loss.

Test result

	INT8	FP32
Accuracy (eval-accuracy)	0.7838	0.7915
Model size (MB)	133	418

Load with optimum:

from optimum.intel import INCModelForMultipleChoice

model_id = "Intel/bert-base-uncased-finetuned-swag-int8-static"
int8_model = INCModelForMultipleChoice.from_pretrained(model_id)