README.md

metadata

language:
  - en
license: mit
tags:
  - text-classfication
  - int8
  - Intel® Neural Compressor
  - neural-compressor
  - PostTrainingDynamic
  - onnx
datasets:
  - nyu-mll/glue
metrics:
  - f1
model-index:
  - name: camembert-base-mrpc-int8-dynamic
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: GLUE MRPC
          type: glue
          args: mrpc
        metrics:
          - type: f1
            value: 0.8842832469775476
            name: F1

INT8 camembert-base-mrpc

Post-training dynamic quantization

PyTorch

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

The original fp32 model comes from the fine-tuned model camembert-base-mrpc.

The linear module roberta.encoder.layer.6.attention.self.query falls back to fp32 to meet the 1% relative accuracy loss.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.8843	0.8928
Model size (MB)	180	422

Load with Intel® Neural Compressor:

from optimum.intel import INCModelForSequenceClassification

model_id = "Intel/camembert-base-mrpc-int8-dynamic"
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 camembert-base-mrpc.

Test result

	INT8	FP32
Accuracy (eval-f1)	0.8819	0.8928
Model size (MB)	113	423

Load ONNX model:

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