Tapex Large Finetuned Tabfact

🚀 TAPEX (large-sized model)

TAPEX is a pre - training approach that endows existing models with table reasoning skills. It was proposed in a research paper and has a specific fine - tuned version for table fact verification.

✨ Features

• TAPEX is a conceptually simple and empirically powerful pre - training approach for table reasoning.
• Based on the BART architecture, it combines a bidirectional encoder and an autoregressive decoder.
• This fine - tuned model can be used for table fact verification.

📦 Installation

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

💻 Usage Examples

Basic Usage

from transformers import TapexTokenizer, BartForSequenceClassification
import pandas as pd

tokenizer = TapexTokenizer.from_pretrained("microsoft/tapex-large-finetuned-tabfact")
model = BartForSequenceClassification.from_pretrained("microsoft/tapex-large-finetuned-tabfact")

data = {
    "year": [1896, 1900, 1904, 2004, 2008, 2012],
    "city": ["athens", "paris", "st. louis", "athens", "beijing", "london"]
}
table = pd.DataFrame.from_dict(data)

# tapex accepts uncased input since it is pre-trained on the uncased corpus
query = "beijing hosts the olympic games in 2012"
encoding = tokenizer(table=table, query=query, return_tensors="pt")

outputs = model(**encoding)
output_id = int(outputs.logits[0].argmax(dim=0))
print(model.config.id2label[output_id])
# Refused

Advanced Usage

No advanced usage code examples are provided in the original document, so this part is skipped.

📚 Documentation

TAPEX (Table Pre - training via Execution) is a pre - training approach to empower existing models with table reasoning skills. It realizes table pre - training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries.

This model is the tapex - base model fine - tuned on the Tabfact dataset. You can use the model for table fact verification.

How to Eval

Please find the eval script here.

🔧 Technical Details

TAPEX is based on the BART architecture, which is a transformer encoder - encoder (seq2seq) model. It has a bidirectional (BERT - like) encoder and an autoregressive (GPT - like) decoder.

📄 License

This model is licensed under the MIT license.

BibTeX entry and citation info

@inproceedings{
    liu2022tapex,
    title={{TAPEX}: Table Pre-training via Learning a Neural {SQL} Executor},
    author={Qian Liu and Bei Chen and Jiaqi Guo and Morteza Ziyadi and Zeqi Lin and Weizhu Chen and Jian-Guang Lou},
    booktitle={International Conference on Learning Representations},
    year={2022},
    url={https://openreview.net/forum?id=O50443AsCP}
}