datasets:
- xnli
model-index:
- name: multilingual-e5-large-xnli
results: []
pipeline_tag: zero-shot-classification
language:
- multilingual
- en
- ar
- bg
- de
- el
- es
- fr
- ru
- sw
- th
- tr
- ur
- vi
- zh
license: mit
widget:
- text: Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU
candidate_labels: politics, economy, entertainment, environment
multilingual-e5-large-xnli
This model is a fine-tuned version of intfloat/multilingual-e5-large on the XNLI dataset.
Model description
Text Embeddings by Weakly-Supervised Contrastive Pre-training. 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:
from transformers import pipeline
classifier = pipeline("zero-shot-classification",
model="mjwong/multilingual-e5-large-xnli")
You can then use this pipeline to classify sequences into any of the class names you specify.
sequence_to_classify = "Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU"
candidate_labels = ["politics", "economy", "entertainment", "environment"]
classifier(sequence_to_classify, candidate_labels)
#{'sequence': 'Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU',
# 'labels': ['politics', 'economy', 'entertainment', 'environment'],
# 'scores': [0.6869393587112427,
# 0.18112628161907196,
# 0.07022464275360107,
# 0.06170979142189026]}
If more than one candidate label can be correct, pass multi_class=True to calculate each class independently:
candidate_labels = ["politics", "economy", "entertainment", "environment"]
classifier(sequence_to_classify, candidate_labels, multi_label=True)
#{'sequence': 'Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU',
# 'labels': ['politics', 'economy', 'entertainment', 'environment'],
# 'scores': [0.9068101644515991,
# 0.2646591067314148,
# 0.00299322628416121,
# 0.0016029390972107649]}
Eval results
The model was evaluated using the XNLI test sets on 14 languages: English (en), Arabic (ar), Bulgarian (bg), German (de), Greek (el), Spanish (es), French (fr), 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 | ru | sw | th | tr | ur | vi | zh |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| multilingual-e5-base-xnli | 0.849 | 0.771 | 0.800 | 0.796 | 0.795 | 0.812 | 0.801 | 0.783 | 0.731 | 0.767 | 0.771 | 0.710 | 0.789 | 0.786 |
| multilingual-e5-large-xnli | 0.867 | 0.798 | 0.829 | 0.821 | 0.820 | 0.838 | 0.828 | 0.810 | 0.752 | 0.787 | 0.794 | 0.726 | 0.804 | 0.810 |
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: 1
Framework versions
- Transformers 4.28.1
- Pytorch 1.12.1+cu116
- Datasets 2.11.0
- Tokenizers 0.12.1