Xlm Roberta Large Xquad

🚀 xlm-roberta-large for multilingual QA

This project utilizes the xlm - roberta - large model for multilingual extractive question - answering tasks, trained and tested on the XQuAD dataset.

🚀 Quick Start

This model is designed for multilingual extractive question - answering. It's trained and tested on the XQuAD dataset. You can quickly start using it through the provided code examples in different frameworks.

✨ Features

• Multilingual Support: Capable of handling extractive question - answering tasks in multiple languages.
• High Performance: Achieves good results on the XQuAD test set, with an exact match of 87.12546816479401 and an F1 score of 94.77703248802527.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

In Transformers

from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline

model_name = "alon-albalak/xlm-roberta-large-xquad"

# a) Get predictions
nlp = pipeline('question-answering', model=model_name, tokenizer=model_name)
QA_input = {
    'question': 'Why is model conversion important?',
    'context': 'The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks.'
}
res = nlp(QA_input)

# b) Load model & tokenizer
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

In FARM

from farm.modeling.adaptive_model import AdaptiveModel
from farm.modeling.tokenization import Tokenizer
from farm.infer import QAInferencer

model_name = "alon-albalak/xlm-roberta-large-xquad"

# a) Get predictions
nlp = QAInferencer.load(model_name)
QA_input = [{"questions": ["Why is model conversion important?"],
             "text": "The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks."}]
res = nlp.inference_from_dicts(dicts=QA_input, rest_api_schema=True)

# b) Load model & tokenizer
model = AdaptiveModel.convert_from_transformers(model_name, device="cpu", task_type="question_answering")
tokenizer = Tokenizer.load(model_name)

In Haystack

reader = FARMReader(model_name_or_path="alon-albalak/xlm-roberta-large-xquad")
# or 
reader = TransformersReader(model="alon-albalak/xlm-roberta-large-xquad",tokenizer="alon-albalak/xlm-roberta-large-xquad")

Usage instructions for FARM and Haystack were adopted from https://huggingface.co/deepset/xlm-roberta-large-squad2

📚 Documentation

Model Information

Hyperparameters

batch_size = 48
n_epochs = 13
max_seq_len = 384
doc_stride = 128
learning_rate = 3e-5

Performance

Evaluated on held - out test set from XQuAD

"exact_match": 87.12546816479401,
"f1": 94.77703248802527,
"test_samples": 2307