File size: 994 Bytes
719ffc1 8858d8a 928ebf2 f7c4dfd 8858d8a 719ffc1 8858d8a f7c4dfd 8858d8a f7c4dfd 8858d8a 2f02c5e 8858d8a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 |
---
license: apache-2.0
tags:
- int8
- Intel® Neural Compressor
- neural-compressor
- PostTrainingDynamic
datasets:
- mnli
metrics:
- accuracy
---
# INT8 T5 small finetuned on XSum
### Post-training dynamic quantization
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 [adasnew/t5-small-xsum](https://huggingface.co/adasnew/t5-small-xsum).
The linear modules **lm.head**, fall back to fp32 for less than 1% relative accuracy loss.
### Evaluation result
| |INT8|FP32|
|---|:---:|:---:|
| **Accuracy (eval-rouge1)** | 29.9008 |29.9592|
| **Model size** |154M|242M|
### Load with optimum:
```python
from optimum.intel import INCModelForSeq2SeqLM
model_id = "Intel/t5-small-xsum-int8-dynamic-inc"
int8_model = INCModelForSeq2SeqLM.from_pretrained(model_id)
```
|