tabred

Tabular model benchmark

An effort to analyze pitfalls and improve the evaluation of tabular deep learning models by creating a new benchmark with real-world datasets

TabReD: Analyzing Pitfalls and Filling the Gaps in Tabular Deep Learning Benchmarks

GitHub

57 stars
3 watching
2 forks
Language: Python
last commit: almost 2 years ago
benchmarksmachine-learningresearchtabular-data

Related projects:

RepositoryDescriptionStars
yandex/repA toolset for building and running reproducible machine learning experiments in Python689
manujosephv/pytorch_tabularA deep learning framework specifically designed for tabular data, providing a standardized approach to modeling and deploying complex machine learning models.1,404
juliadata/dataframes.jlA package providing tools and data structures for efficiently working with tabular data in Julia.1,738
yandexdataschool/ysda_deeplearning17Repository containing lecture and seminar materials for a deep learning course taught in 2017116
datonic/datadexA platform for collaborative open data management and analysis264
rajasegar/htmx-tabularAn application that demonstrates tabular data display with features like search, sorting, and pagination using the htmx library in Node.8
hepdata/hepdataA web application for managing and sharing high-energy physics data from experiments41
pyg-team/pytorch-frameA deep learning framework for handling heterogeneous tabular data with diverse column types582
awesomedata/awesome-public-datasetsA curated collection of high-quality public datasets organized by topic.61,377
mohamedadaly/labrA dataset of Arabic book reviews for natural language processing tasks44
rucaibox/recsysdatasetsA repository of public data sources for Recommender Systems.887
cidree/forestdataA package providing easy access to forestry and land use datasets.13
jxshin/mzdataA comprehensive dataset of Mozilla issue tracking history, providing multiple extracts and levels for analysis.7
gopherdata/resourcesA collection of Go-based resources and tools for data science tasks879
datasciencemasters/dataA curated list of accessible data sources with clear licensing terms and moderate restrictions.506