File size: 2,639 Bytes
586a45b
fde7023
 
 
 
 
 
 
 
586a45b
 
fde7023
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e12b9e3
d92ed9c
e12b9e3
4ee38d0
e12b9e3
 
 
 
 
fde7023
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
datasets:
- glue
model-index:
- name: e5-large-mnli
  results: []
pipeline_tag: zero-shot-classification
language:
- en
license: mit
---

# e5-large-mnli

This model is a fine-tuned version of [intfloat/e5-large](https://huggingface.co/intfloat/e5-large) on the glue dataset.

## Model description

[Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://arxiv.org/pdf/2212.03533.pdf).
Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, Furu Wei, arXiv 2022

## How to use the model

The model can be loaded with the `zero-shot-classification` pipeline like so:

```python
from transformers import pipeline
classifier = pipeline("zero-shot-classification",
                      model="mjwong/e5-large-mnli")
```

You can then use this pipeline to classify sequences into any of the class names you specify.

```python
sequence_to_classify = "one day I will see the world"
candidate_labels = ['travel', 'cooking', 'dancing']
classifier(sequence_to_classify, candidate_labels)
#{'sequence': 'one day I will see the world',
# 'labels': ['travel', 'dancing', 'cooking'],
# 'scores': [0.9494319558143616, 0.044598229229450226, 0.00596982054412365]}
```

If more than one candidate label can be correct, pass `multi_class=True` to calculate each class independently:

```python
candidate_labels = ['travel', 'cooking', 'dancing', 'exploration']
classifier(sequence_to_classify, candidate_labels, multi_class=True)
#{'sequence': 'one day I will see the world',
# 'labels': ['exploration', 'travel', 'dancing', 'cooking'],
# 'scores': [0.9918234944343567,
#  0.9867327213287354,
#  0.40335655212402344,
#  0.0020157278049737215]}
```

### Eval results
The model was evaluated using the dev sets for MultiNLI and test sets for ANLI. The metric used is accuracy.

|Datasets|mnli_dev_m|mnli_dev_mm|anli_test_r1|anli_test_r2|anli_test_r3|
| :---: | :---: | :---: | :---: | :---: | :---: |
|[e5-base-mnli](https://huggingface.co/mjwong/e5-base-mnli)|0.840|0.839|0.231|0.285|0.309|
|[e5-large-mnli](https://huggingface.co/mjwong/e5-large-mnli)|0.868|0.869|0.301|0.296|0.294|
|[e5-large-mnli-anli](https://huggingface.co/mjwong/e5-large-mnli-anli)|0.843|0.848|0.646|0.484|0.458|

### Training hyperparameters

The following hyperparameters were used during training:

- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2

### Framework versions
- Transformers 4.28.1
- Pytorch 1.12.1+cu116
- Datasets 2.11.0
- Tokenizers 0.12.1