README.md

metadata

language: en
license: mit
tags:
  - text-classfication
  - int8
  - Intel® Neural Compressor
  - PostTrainingDynamic
  - onnx
datasets:
  - mrpc
metrics:
  - f1

INT8 xlm-roberta base finetuned MRPC

Post-training dynamic quantization

ONNX

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

The original fp32 model comes from the fine-tuned model Intel/xlm-roberta-base-mrpc.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.8966	0.9010
Model size (MB)	354	1061

Load ONNX model:

from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/xlm-roberta-base-mrpc-int8-dynamic')