tapex large finetuned tabfact

提供商microsoft
分类table-question-answering
许可证mit
下载量163
星标0

简介

TaPEX Large Finetuned TabFact 是微软推出的专门针对表格问答（Table QA）优化的模型。不同于通用大模型在处理复杂表格时容易产生的“幻觉”或对行列对应关系理解偏差，TaPEX 经过 TabFact 数据集的深度微调，核心能力在于能够精准地对表格内容进行事实核查和逻辑推理。对于开发者而言，它非常适合集成到自动化报表分析、数据校验或结构化知识库查询场景中。上手难度较低，可作为轻量级的专用插件，与 Pandas 等数据处理工具配合，高效完成对表格事实的真伪判定。

核心亮点

  • 专注表格事实核查，逻辑推理精准度高
  • 有效解决通用模型处理表格时的幻觉问题
  • 适用于自动化报表审计与结构化数据问答
  • MIT 协议开源，易于集成到企业私有化链路

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("microsoft/tapex-large-finetuned-tabfact")
tokenizer = AutoTokenizer.from_pretrained("microsoft/tapex-large-finetuned-tabfact")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download microsoft/tapex-large-finetuned-tabfact

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download microsoft/tapex-large-finetuned-tabfact config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('microsoft/tapex-large-finetuned-tabfact')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/microsoft/tapex-large-finetuned-tabfact

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/microsoft/tapex-large-finetuned-tabfact

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('microsoft/tapex-large-finetuned-tabfact')
tokenizer = AutoTokenizer.from_pretrained('microsoft/tapex-large-finetuned-tabfact')

模型下载

我们推荐使用命令行或者 ModelScope SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 ModelScope：

操作指引 
pip install modelscope

命令行下载

下载完整模型库

下载完整模型库 
modelscope download --model microsoft/tapex-large-finetuned-tabfact

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例）

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例） 
modelscope download --model microsoft/tapex-large-finetuned-tabfact README.md --local_dir ./dir

更多更丰富的命令行下载选项，可参见具体文档

SDK 下载

SDK 下载 
# 模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('microsoft/tapex-large-finetuned-tabfact')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://www.modelscope.cn/microsoft/tapex-large-finetuned-tabfact.git

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://www.modelscope.cn/microsoft/tapex-large-finetuned-tabfact.git

ModelScope 模型页直接下载模型文件；无需将模型文件放在本站服务器。

Notebook 快速开发

下载并安装 ModelScope library

下载并安装 ModelScope library 
pip install "modelscope[audio,cv,nlp,multi-modal,science]" -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html

模型加载和推理

模型加载和推理 
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks

p = pipeline('text-generation', 'microsoft/tapex-large-finetuned-tabfact')

完整文档

来源: HuggingFace

---
language: en
tags:

  • tapex

  • table-question-answering

datasets:
  • tab_fact

license: mit
---

TAPEX (large-sized model)

TAPEX was proposed in TAPEX: Table Pre-training via Learning a Neural SQL Executor by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou. The original repo can be found here.

Model description

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

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

This model is the tapex-base model fine-tuned on the Tabfact dataset.

Intended Uses

You can use the model for table fact verficiation.

How to Use

Here is how to use this model in transformers:

python
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

How to Eval

Please find the eval script here.

BibTeX entry and citation info

bibtex
@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}
}
查看官方来源