File size: 2,280 Bytes
0483abe
bb650cc
 
0483abe
4719958
 
 
cd77fc8
4719958
3518c77
bb650cc
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0483abe
bb650cc
 
 
3518c77
 
 
bb650cc
ac02df7
bb650cc
 
 
 
 
 
3518c77
bb650cc
 
 
 
 
 
3518c77
bb650cc
 
ac02df7
 
bb650cc
 
 
3518c77
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
---
language:
- en
license: mit
tags:
- text-classfication
- int8
- Intel® Neural Compressor
- PostTrainingStatic
- onnx
datasets:
- glue
metrics:
- f1
model-index:
- name: electra-small-discriminator-mrpc-int8-static
  results:
  - task:
      name: Text Classification
      type: text-classification
    dataset:
      name: GLUE MRPC
      type: glue
      args: mrpc
    metrics:
    - name: F1
      type: f1
      value: 0.900709219858156
---

# INT8 electra-small-discriminator-mrpc

##  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 [electra-small-discriminator-mrpc](https://huggingface.co/Intel/electra-small-discriminator-mrpc).

The calibration dataloader is the train dataloader. The default calibration sampling size 300 isn't divisible exactly by batch size 8, so
 the real sampling size is 304.

#### Test result

|   |INT8|FP32|
|---|:---:|:---:|
| **Accuracy (eval-f1)** |0.9007|0.8983|
| **Model size (MB)**  |14|51.8|

#### Load with optimum:

```python
from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSequenceClassification
int8_model = IncQuantizedModelForSequenceClassification.from_pretrained(
    'Intel/electra-small-discriminator-mrpc-int8-static',
)
```

### ONNX

This is an INT8 ONNX model quantized with [Intel® Neural Compressor](https://github.com/intel/neural-compressor).

The original fp32 model comes from the fine-tuned model [electra-small-discriminator-mrpc](https://huggingface.co/Intel/electra-small-discriminator-mrpc).

The calibration dataloader is the eval dataloader. The default calibration sampling size 100 isn't divisible exactly by batch size 8. So the real sampling size is 104.

#### Test result

|   |INT8|FP32|
|---|:---:|:---:|
| **Accuracy (eval-f1)** |0.8993|0.8983|
| **Model size (MB)**  |32|52|


#### Load ONNX model:

```python
from optimum.onnxruntime import ORTModelForSequenceClassification
model = ORTModelForSequenceClassification.from_pretrained('Intel/electra-small-discriminator-mrpc-int8-static')
```