Dynamically quantized DistilBERT base uncased finetuned SST-2
Table of Contents
Model Details
Model Description: This model is a DistilBERT fine-tuned on SST-2 dynamically quantized with optimum-intel through the usage of huggingface/optimum-intel through the usage of Intel® Neural Compressor.
- Model Type: Text Classification
- Language(s): English
- License: Apache-2.0
- Parent Model: For more details on the original model, we encourage users to check out this model card.
How to Get Started With the Model
PyTorch
To load the quantized model, you can do as follows:
from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSequenceClassification
model = IncQuantizedModelForSequenceClassification.from_pretrained("Intel/distilbert-base-uncased-finetuned-sst-2-english-int8-dynamic")
ONNX
This is an INT8 ONNX model quantized with Intel® Neural Compressor.
The original fp32 model comes from the fine-tuned model DistilBERT.
Test result
| INT8 | FP32 | |
|---|---|---|
| Accuracy (eval-accuracy) | 0.9025 | 0.9106 |
| Model size (MB) | 165 | 256 |
Load ONNX model:
from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/distilbert-base-uncased-finetuned-sst-2-english-int8-dynamic')
