|
--- |
|
language: en |
|
license: apache-2.0 |
|
tags: |
|
- text-classfication |
|
- int8 |
|
- Intel® Neural Compressor |
|
- PostTrainingStatic |
|
- bert |
|
datasets: |
|
- mrpc |
|
- sst2 |
|
metrics: |
|
- f1 |
|
--- |
|
|
|
# INT8 BERT base uncased finetuned sst2 |
|
|
|
## Post-training static quantization |
|
|
|
### PyTorch |
|
|
|
This is an INT8 PyTorch model quantized with [huggingface/optimum-intel](https://github.com/huggingface/optimum-intel) through the usage of [Intel® Neural Compressor](https://github.com/intel/neural-compressor). |
|
|
|
The original fp32 model comes from the fine-tuned model [gchhablani/bert-base-cased-finetuned-sst2](https://huggingface.co/gchhablani/bert-base-cased-finetuned-sst2). |
|
|
|
#### Test result |
|
|
|
| |INT8|FP32| |
|
|---|:---:|:---:| |
|
| **Accuracy (eval-f1)** |0.9151|0.9232| |
|
| **Model size (MB)** |111|433| |
|
|
|
#### Load with optimum: |
|
|
|
```python |
|
from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSequenceClassification |
|
int8_model = IncQuantizedModelForSequenceClassification.from_pretrained( |
|
'Intel/bert-base-cased-finetuned-sst2-int8', |
|
) |
|
``` |
