File size: 4,559 Bytes
53825d0 a3060a2 53825d0 90484af 53825d0 1c8f49d 53825d0 bde5676 53825d0 bde5676 53825d0 bde5676 53825d0 bde5676 53825d0 a3060a2 |
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 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 |
---
language:
- multilingual
- en
- ar
- bg
- de
- el
- es
- fr
- hi
- ru
- sw
- th
- tr
- ur
- vi
- zh
license: apache-2.0
datasets:
- xnli
pipeline_tag: zero-shot-classification
base_model: Alibaba-NLP/gte-multilingual-base
model-index:
- name: gte-multilingual-base-xnli
results: []
---
# gte-multilingual-base-xnli
This model is a fine-tuned version of [Alibaba-NLP/gte-multilingual-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-base) on the XNLI dataset.
## Model description
[mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval](https://arxiv.org/pdf/2407.19669).
Xin Zhang, Yanzhao Zhang, Dingkun Long, Wen Xie, Ziqi Dai, Jialong Tang, Huan Lin, Baosong Yang, Pengjun Xie, Fei Huang, Meishan Zhang, Wenjie Li, Min Zhang, arXiv 2024
## How to use the model
### With the zero-shot classification pipeline
The model can be loaded with the `zero-shot-classification` pipeline like so:
```python
from transformers import AutoTokenizer, pipeline
model = "mjwong/gte-multilingual-base-xnli"
tokenizer = AutoTokenizer.from_pretrained(model)
classifier = pipeline("zero-shot-classification",
model=model,
tokenizer=tokenizer,
trust_remote_code=True
)
```
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)
```
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)
```
### With manual PyTorch
The model can also be applied on NLI tasks like so:
```python
import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification
# device = "cuda:0" or "cpu"
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
model_name = "mjwong/gte-multilingual-base-xnli"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name, trust_remote_code=True)
premise = "But I thought you'd sworn off coffee."
hypothesis = "I thought that you vowed to drink more coffee."
input = tokenizer(premise, hypothesis, truncation=True, return_tensors="pt")
output = model(input["input_ids"].to(device))
prediction = torch.softmax(output["logits"][0], -1).tolist()
label_names = ["entailment", "neutral", "contradiction"]
prediction = {name: round(float(pred) * 100, 2) for pred, name in zip(prediction, label_names)}
print(prediction)
```
### Eval results
The model was evaluated using the XNLI test sets on 15 languages: English (en), Arabic (ar), Bulgarian (bg), German (de), Greek (el), Spanish (es), French (fr), Hindi (hi), Russian (ru), Swahili (sw), Thai (th), Turkish (tr), Urdu (ur), Vietnam (vi) and Chinese (zh). The metric used is accuracy.
|Datasets|en|ar|bg|de|el|es|fr|hi|ru|sw|th|tr|ur|vi|zh|
| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
|[gte-multilingual-base-xnli](https://huggingface.co/mjwong/gte-multilingual-base-xnli)|0.854|0.767|0.811|0.798|0.801|0.820|0.818|0.753|0.792|0.719|0.766|0.769|0.701|0.799|0.798|
|[gte-multilingual-base-xnli-anli](https://huggingface.co/mjwong/gte-multilingual-base-xnli-anli)|0.843|0.738|0.793|0.773|0.776|0.801|0.788|0.727|0.775|0.689|0.746|0.747|0.687|0.773|0.779|
The model was also 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|
| :---: | :---: | :---: | :---: | :---: | :---: |
|[gte-multilingual-base-xnli](https://huggingface.co/mjwong/gte-multilingual-base-xnli)|0.852|0.852|0.295|0.292|0.336|
|[gte-multilingual-base-xnli-anli](https://huggingface.co/mjwong/gte-multilingual-base-xnli-anli)|0.834|0.837|0.567|0.445|0.443|
### 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
### Framework versions
- Transformers 4.41.0
- Pytorch 2.6.0+cu124
- Datasets 3.2.0
- Tokenizers 0.19.1 |