README.md

metadata

license: mit
tags:
  - text-classfication
  - int8
  - Intel® Neural Compressor
  - neural-compressor
  - PostTrainingStatic
  - onnx
datasets:
  - glue
metrics:
  - f1

INT8 deberta-v3-base-mrpc

Post-training static quantization

ONNX

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

The original fp32 model comes from the fine-tuned model Intel/deberta-v3-base-mrpc.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.9185	0.9223
Model size (MB)	361	705

Load ONNX model:

from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/deberta-v3-base-mrpc-int8-static')