|
--- |
|
language: en |
|
license: apache-2.0 |
|
tags: |
|
- text-classfication |
|
- int8 |
|
- QuantizationAwareTraining |
|
datasets: |
|
- mrpc |
|
metrics: |
|
- f1 |
|
--- |
|
|
|
# INT8 BERT base uncased finetuned MRPC |
|
|
|
### QuantizationAwareTraining |
|
|
|
This is an INT8 PyTorch model quantized with [Intel® Neural Compressor](https://github.com/intel/neural-compressor). |
|
|
|
The original fp32 model comes from the fine-tuned model [Intel/bert-base-uncased-mrpc](https://huggingface.co/Intel/bert-base-uncased-mrpc). |
|
|
|
### Test result |
|
|
|
- Batch size = 8 |
|
- [Amazon Web Services](https://aws.amazon.com/) c6i.xlarge (Intel ICE Lake: 4 vCPUs, 8g Memory) instance. |
|
|
|
| |INT8|FP32| |
|
|---|:---:|:---:| |
|
| **Throughput (samples/sec)** |24.263|11.202| |
|
| **Accuracy (eval-f1)** |0.9153|0.9042| |
|
| **Model size (MB)** |174|418| |
|
|
|
### Load with Intel® Neural Compressor (build from source): |
|
|
|
```python |
|
from neural_compressor.utils.load_huggingface import OptimizedModel |
|
int8_model = OptimizedModel.from_pretrained( |
|
'Intel/bert-base-uncased-mrpc-int8-qat', |
|
) |
|
``` |
|
|
|
Notes: |
|
- The INT8 model has better performance than the FP32 model when the CPU is fully occupied. Otherwise, there will be the illusion that INT8 is inferior to FP32. |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- eval_steps: 100 |
|
- load_best_model_at_end: True |
|
- metric_for_best_model: f1 |
|
- early_stopping_patience = 6 |
|
- early_stopping_threshold = 0.001 |
|
|
