Title: MISS: An Incomplete Tabular Data Representation System with Missing Mechanism Learning

This project was edited by Lzy from July to August 2025, when Lzy was at Zhejiang University

Updated

• requirements.txt
• README.md
• scripts/tabular.sh
• data
• dataset
• result
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/train4.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/train6_lzy.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/rtdl_our.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/model_comparison.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/IPS.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/lib/results_statistics.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/lib/make_info_json.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/lib/make_divided_npy.py
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/exp/transformer-q-lr/HI
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/exp/transformer-q-lr/News
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/exp/transformer-q-lr/temperature
• code/tabular/tabular/tabular-dl-num-embeddings-main-our/exp/transformer-q-lr/gas

Introduction

This folder (i.e., ./code) holds the source codes of the proposed system MISS on top of the four transformer-based models, including

• MISS-saint (in ./coda/saint/saint-main)
• MISS-tabular (in ./coda/tabular/tabular/tabular-dl-num-embeddings-main-our)
• MISS-tabformer (in ./coda/representation/TabSurvey/TabSurvey-main)
• MISS-npt (in ./code/non-parametric-transformers-main/non-parametric-transformers-main-our)

Requirements

To install requirements:

pip install -r requirements.txt

Dataset Description

All of the original datasets can refer to the links:

• News: https://archive.ics.uci.edu/dataset/332/online+news+popularity
• HI: https://www.openml.org/search?type=data&status=active&id=23512&sort=runs
• gas: https://archive.ics.uci.edu/dataset/322/gas+sensor+array+under+dynamic+gas+mixtures
• temperature: https://archive.ics.uci.edu/dataset/514/bias+correction+of+numerical+prediction+model+temperature+forecast

Usage

• You can use scripts in /scripts to run all the experiments