Multilingual E5 Large Xnli

🚀 multilingual-e5-large-xnli

This model is a fine - tuned version of intfloat/multilingual-e5-large on the XNLI dataset, enabling zero - shot classification across multiple languages.

🚀 Quick Start

This model can be used for zero - shot classification tasks. You can load the model with the zero - shot - classification pipeline or use it manually with PyTorch.

✨ Features

• Multilingual Support: Supports multiple languages such as English, Arabic, Bulgarian, etc.
• Zero - Shot Classification: Can classify sequences into user - specified classes without prior training on those classes.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

With the zero - shot classification pipeline

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

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)

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)

With manual PyTorch

The model can also be applied on NLI tasks:

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/multilingual - e5 - large - xnli"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

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)

📚 Documentation

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.

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

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

🔧 Technical Details

The model is based on the pre - trained model [intfloat/multilingual - e5 - large](https://huggingface.co/intfloat/multilingual - e5 - large) and fine - tuned on the XNLI dataset. It uses techniques from Text Embeddings by Weakly - Supervised Contrastive Pre - training.

📄 License

This project is licensed under the MIT license.