README.md

metadata

language:
  - en
license: mit
tags:
  - text-classfication
  - int8
  - Intel® Neural Compressor
  - PostTrainingStatic
  - onnx
datasets:
  - glue
metrics:
  - f1
model-index:
  - name: electra-small-discriminator-mrpc-int8-static
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE MRPC
          type: glue
          args: mrpc
        metrics:
          - name: F1
            type: f1
            value: 0.900709219858156

INT8 electra-small-discriminator-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 electra-small-discriminator-mrpc.

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.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.9007	0.8983
Model size (MB)	14	51.8

Load with optimum:

from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSequenceClassification
int8_model = IncQuantizedModelForSequenceClassification.from_pretrained(
    'Intel/electra-small-discriminator-mrpc-int8-static',
)

ONNX

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

The original fp32 model comes from the fine-tuned model electra-small-discriminator-mrpc.

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

Test result

	INT8	FP32
Accuracy (eval-f1)	0.8993	0.8983
Model size (MB)	32	52

Load ONNX model:

from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/electra-small-discriminator-mrpc-int8-static')