Intel/MiniLM-L12-H384-uncased-mrpc-int8-static-inc

INT8 MiniLM-L12-H384 finetuned MRPC

Post-training static quantization

PyTorch

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 Intel/MiniLM-L12-H384-uncased-mrpc.

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 module bert.encoder.layer.6.attention.self.key falls back to fp32 to meet the 1% relative accuracy loss.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.9039	0.9097
Model size (MB)	33.5	127

Load with optimum:

from optimum.intel import INCModelForSequenceClassification 

model_id = "Intel/MiniLM-L12-H384-uncased-mrpc-int8-static"
int8_model = INCModelForSequenceClassification.from_pretrained(model_id)

ONNX

This is an INT8 ONNX model quantized with Intel® Neural Compressor.

The original fp32 model comes from the fine-tuned model Intel/MiniLM-L12-H384-uncased-mrpc.

The calibration dataloader is the eval dataloader. The calibration sampling size is 100.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.9013	0.9097
Model size (MB)	33	128

Load ONNX model:

from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/MiniLM-L12-H384-uncased-mrpc-int8-static')