Bert Base Japanese V3 Nli Jsnli

🚀 Cross-Encoder for Natural Language Inference (NLI) for Japanese

This is a cross-encoder model for natural language inference (NLI) in Japanese, trained on specific datasets and offering multiple usage scenarios.

🚀 Quick Start

Considering the results of the JNLI evaluation result, we recommend using akiFQC/bert-base-japanese-v3_nli-jsnli-jnli-jsick for natural language inference in Japanese.

✨ Features

• This model was trained using SentenceTransformers Cross-Encoder class.
• This model is based on tohoku-nlp/bert-base-japanese-v3.

📦 Installation

No specific installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

Pre-trained models can be used like this:

from sentence_transformers import CrossEncoder
model = CrossEncoder('akiFQC/bert-base-japanese-v3_nli-jsnli')
scores = model.predict([('男はピザを食べています', '男は何かを食べています'), ('黒いレーシングカーが観衆の前から発車します。', '男は誰もいない道を運転しています。')])

#Convert scores to labels
label_mapping = ['entailment', 'neutral', 'contradiction',]
labels = [label_mapping[score_max] for score_max in scores.argmax(axis=1)]

Advanced Usage - Usage with Transformers AutoModel

You can use the model also directly with Transformers library (without SentenceTransformers library):

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model = AutoModelForSequenceClassification.from_pretrained('cross-encoder/nli-deberta-v3-base')
tokenizer = AutoTokenizer.from_pretrained('cross-encoder/nli-deberta-v3-base')

features = tokenizer(['男はピザを食べています', '黒いレーシングカーが観衆の前から発車します。'], ['男は何かを食べています', '男は誰もいない道を運転しています。'],  padding=True, truncation=True, return_tensors="pt")

model.eval()
with torch.no_grad():
    scores = model(**features).logits
    label_mapping = ['contradiction', 'entailment', 'neutral']
    labels = [label_mapping[score_max] for score_max in scores.argmax(dim=1)]
    print(labels)

Advanced Usage - Zero-Shot Classification

This model can also be used for zero-shot-classification:

from transformers import pipeline

classifier = pipeline("zero-shot-classification", model='akiFQC/bert-base-japanese-v3_nli-jsnli')

sent = "Appleは先程、iPhoneの最新機種について発表しました。"
candidate_labels = ["技術", "スポーツ", "政治"]
res = classifier(sent, candidate_labels)
print(res)

📚 Documentation

Training Data

The model was trained on the JSNLI datasets. For a given sentence pair, it will output three scores corresponding to the labels: {0:"entailment", 1:"neutral", 2:"contradiction}.

Benchmarks

JGLUE - JNLI validation set accuracy: 0.770

📄 License

This project is licensed under the CC-BY-SA-4.0 license.