Datasets:
task_id stringlengths 13 13 | X listlengths 1k 1k | y listlengths 1k 1k | n_features int32 100 100 | mb_mask listlengths 100 100 | mb_ratio float32 0.13 0.9 | density float64 0.2 0.41 | dag dict | scm dict | target dict | meta dict |
|---|---|---|---|---|---|---|---|---|---|---|
task_00683b62 | [[-1.0779759883880615,-0.6852225065231323,1.3463866710662842,0.7228516340255737,-0.4921930134296417,(...TRUNCATED) | [-1.378145456314087,0.004080341197550297,1.8120592832565308,0.9082462787628174,-0.14349448680877686,(...TRUNCATED) | 100 | [false,false,false,false,true,false,false,false,false,false,false,false,false,true,false,false,true,(...TRUNCATED) | 0.21 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,64],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_GAUSSIAN","noise_model":"GAUSSIAN","pnl":false,"coeff_range":1.0,"noise_std":0.5(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,true,false,false,false,false,false,false,f(...TRUNCATED) | {
"n_samples": 1000,
"seed": 1249693216
} |
task_3c07c91b | [[-0.6577381491661072,-0.4469158947467804,1.0436208248138428,-0.7493091225624084,-1.651067852973938,(...TRUNCATED) | [-0.6824029088020325,0.09647499024868011,1.2255958318710327,0.04425739124417305,0.3502548038959503,1(...TRUNCATED) | 100 | [false,false,false,true,true,false,false,false,false,false,true,false,false,true,false,false,true,tr(...TRUNCATED) | 0.43 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,64],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_GAUSSIAN","noise_model":"GAUSSIAN","pnl":false,"coeff_range":1.0,"noise_std":0.5(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,true,false,false,false,false,false,false,true,fa(...TRUNCATED) | {
"n_samples": 1000,
"seed": 1544123743
} |
task_e220de52 | [[-0.5197003483772278,-2.2590456008911133,0.30848774313926697,-0.02914324961602688,1.580205917358398(...TRUNCATED) | [-1.6353302001953125,0.8600995540618896,0.28282490372657776,-1.0611392259597778,-1.1114835739135742,(...TRUNCATED) | 100 | [true,true,true,true,true,true,false,true,true,true,true,false,true,false,false,false,true,true,true(...TRUNCATED) | 0.78 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,64],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_GAUSSIAN","noise_model":"GAUSSIAN","pnl":false,"coeff_range":1.0,"noise_std":0.5(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,false,false,false,false,false,false,false,(...TRUNCATED) | {
"n_samples": 1000,
"seed": 774196478
} |
task_c4bd8413 | [[1.6596463918685913,0.33345848321914673,-0.4485335946083069,-0.25978097319602966,1.784938931465149,(...TRUNCATED) | [-1.4874531030654907,-0.27010101079940796,-0.7039238214492798,0.40742483735084534,-0.598833262920379(...TRUNCATED) | 100 | [false,false,false,false,true,false,false,false,false,false,false,false,false,true,false,false,true,(...TRUNCATED) | 0.21 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,64],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_GAUSSIAN","noise_model":"GAUSSIAN","pnl":false,"coeff_range":1.0,"noise_std":0.5(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,true,false,false,false,false,false,false,f(...TRUNCATED) | {
"n_samples": 1000,
"seed": 1717614902
} |
task_b9302efe | [[-0.8951002359390259,1.2502763271331787,-0.3618984818458557,-0.0902809277176857,-0.1159962937235832(...TRUNCATED) | [0.19119060039520264,-0.17792242765426636,0.61102694272995,0.38212156295776367,1.6080387830734253,0.(...TRUNCATED) | 100 | [false,false,false,true,true,false,false,false,false,false,true,false,false,true,false,false,true,tr(...TRUNCATED) | 0.43 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,64],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_GAUSSIAN","noise_model":"GAUSSIAN","pnl":false,"coeff_range":1.0,"noise_std":0.5(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,true,false,false,false,false,false,false,true,fa(...TRUNCATED) | {
"n_samples": 1000,
"seed": 1416351476
} |
task_a11add14 | [[-0.620759904384613,0.10411602258682251,-0.2965262830257416,0.39834722876548767,-0.2832464277744293(...TRUNCATED) | [-0.6675601601600647,1.075416922569275,0.5452332496643066,1.0818657875061035,0.4703418016433716,-0.2(...TRUNCATED) | 100 | [false,true,false,true,true,true,false,false,true,true,true,false,true,false,true,false,true,false,f(...TRUNCATED) | 0.64 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,32],[0,43],[0,48],[0,58],[0,63],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_NONGAUSSIAN","noise_model":"LAPLACE","pnl":false,"coeff_range":1.0,"noise_std":0(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,false,true,false,false,false,false,false,f(...TRUNCATED) | {
"n_samples": 1000,
"seed": 154190047
} |
task_e58f021d | [[-0.2669410705566406,0.6181883811950684,-0.7940014600753784,0.7047541737556458,-0.9557840824127197,(...TRUNCATED) | [-0.9408334493637085,-0.5699290633201599,-0.7534362077713013,-0.33249661326408386,-1.522111535072326(...TRUNCATED) | 100 | [false,false,false,false,false,false,false,false,false,false,true,false,true,false,false,false,false(...TRUNCATED) | 0.17 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,44],[0,49],[0,59],[0,63],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_NONGAUSSIAN","noise_model":"LAPLACE","pnl":false,"coeff_range":1.0,"noise_std":0(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,false,false,false,false,false,false,true,f(...TRUNCATED) | {
"n_samples": 1000,
"seed": 159640904
} |
task_e1b890f5 | [[0.11355739831924438,2.6860129833221436,-0.9963576793670654,1.7732889652252197,-0.8615996241569519,(...TRUNCATED) | [-1.0336651802062988,1.0785760879516602,1.3089783191680908,0.5099014043807983,0.7988507747650146,1.8(...TRUNCATED) | 100 | [true,true,false,true,true,true,false,false,true,true,true,true,true,false,true,false,true,true,true(...TRUNCATED) | 0.77 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,43],[0,48],[0,58],[0,63],[0,70],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_NONGAUSSIAN","noise_model":"LAPLACE","pnl":false,"coeff_range":1.0,"noise_std":0(...TRUNCATED) | {"target_node":100,"parent_mask":[true,false,false,false,false,false,false,false,false,false,false,f(...TRUNCATED) | {
"n_samples": 1000,
"seed": 1116185583
} |
task_e2dfed69 | [[0.02249477617442608,-0.8341965079307556,-0.6537241339683533,0.07481981813907623,-0.249811142683029(...TRUNCATED) | [0.42383530735969543,-0.1980883628129959,-0.5778165459632874,-0.45267900824546814,0.3077366054058075(...TRUNCATED) | 100 | [false,true,false,true,false,true,false,true,true,true,true,false,false,false,false,true,true,true,t(...TRUNCATED) | 0.61 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,14],[0,32],[0,43],[0,48],[0,58],[0,63],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_NONGAUSSIAN","noise_model":"LAPLACE","pnl":false,"coeff_range":1.0,"noise_std":0(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,false,false,false,false,false,false,false,(...TRUNCATED) | {
"n_samples": 1000,
"seed": 535973140
} |
task_1aa66334 | [[-1.2449100017547607,-0.4902113080024719,0.04274780675768852,-0.6848593950271606,-0.145239740610122(...TRUNCATED) | [-1.727815866470337,1.0123926401138306,-0.2800917625427246,-0.19611415266990662,-1.25296950340271,-1(...TRUNCATED) | 100 | [false,true,false,false,false,true,false,true,true,true,false,true,false,false,false,false,false,fal(...TRUNCATED) | 0.47 | 0.197822 | {"dag_id":"dag_243003ed","num_nodes":101,"edge_list":[[0,15],[0,33],[0,43],[0,48],[0,58],[0,63],[0,7(...TRUNCATED) | {"scm_type":"LINEAR_NONGAUSSIAN","noise_model":"LAPLACE","pnl":false,"coeff_range":1.0,"noise_std":0(...TRUNCATED) | {"target_node":100,"parent_mask":[false,false,false,false,false,true,false,false,true,false,false,tr(...TRUNCATED) | {
"n_samples": 1000,
"seed": 627155765
} |
SCM3K
Benchmark dataset for the paper:
The Good, the Bad, and the Ugly of Markov Boundary for Tabular Prediction Shu Wan, Abhinav Gorantla, Huan Liu, K. Selçuk Candan
3,450 tabular prediction tasks sampled from random structural causal models (SCMs), totalling 3.45M records (1,000 samples per task). Each task ships with the ground-truth Markov boundary of the target node, so you can evaluate feature selection and prediction under known causal structure. Nine feature-count levels from 40 to 1,000.
Splits
One HF split per feature count F. No predefined train/test
partition — use HF slice syntax (e.g. split="f200[:80%]").
| Split | F |
num_nodes |
DAG density | MB-ratio band | Tasks |
|---|---|---|---|---|---|
| f40 | 40 | 41 | ER [0.2, 0.4] |
[0.10, 0.90] |
300 |
| f60 | 60 | 61 | ER [0.2, 0.4] |
[0.10, 0.90] |
300 |
| f80 | 80 | 81 | ER [0.2, 0.4] |
[0.10, 0.90] |
300 |
| f100 | 100 | 101 | ER [0.2, 0.4] |
[0.10, 0.90] |
300 |
| f200 | 200 | 201 | ER [0.01, 0.02, 0.04] |
[0.05, 0.95] |
450 |
| f400 | 400 | 401 | ER [0.01, 0.02, 0.04] |
[0.05, 0.95] |
450 |
| f600 | 600 | 601 | ER [0.01, 0.02, 0.04] |
[0.05, 0.95] |
450 |
| f800 | 800 | 801 | ER [0.01, 0.02, 0.04] |
[0.05, 0.95] |
450 |
| f1000 | 1000 | 1001 | ER [0.01, 0.02, 0.04] |
[0.05, 0.95] |
450 |
| Total | 3,450 |
Row schema
Each row is one prediction task.
| Field | Type | What it stores |
|---|---|---|
task_id |
string | unique task identifier |
X |
list<list<f32>> | feature matrix, 1,000 x F |
y |
list<f32> | target vector, length 1,000 |
n_features |
int | number of features (= F) |
mb_mask |
list<bool> | true Markov boundary mask over features |
mb_ratio |
float | fraction of features in the Markov boundary |
density |
float | edge density of the generating DAG |
dag |
struct | dag_id, num_nodes, edge_list |
scm |
struct | scm_type, noise_model, pnl, coeff_range, noise_std |
target |
struct | target_node, parent_mask, child_mask, spouse_mask |
meta |
struct | n_samples, seed |
How the data was generated
DAGs: Erdos-Renyi, 5 graphs per (num_nodes, density) pair, seed 42.
SCMs: six families — LINEAR_GAUSSIAN, LINEAR_NONGAUSSIAN,
NL_ANM_GAUSSIAN, NL_ANM_NONGAUSSIAN, PNL, HETEROSKEDASTIC.
Each DAG gets 5 SCM instantiations with n_samples=1000,
coeff_range=1.0, noise_std=0.5.
Quick start
from datasets import load_dataset
ds = load_dataset("CSE472-blanket-challenge/SCM3K", split="f200")
task = ds[0]
X = task["X"] # 1000 x 200
y = task["y"] # 1000
mb = task["mb_mask"] # ground-truth Markov boundary
Citation
@article{wan2026gbu,
title = {The Good, the Bad, and the Ugly of Markov Boundary
for Tabular Prediction},
author = {Wan, Shu and Gorantla, Abhinav and Liu, Huan
and Candan, K. Sel{\c{c}}uk},
year = {2026},
}
- Downloads last month
- 53