Datasets:

TabBioMed
/

maternal_health_risk

row_idx int64 2 380	f_0 float64 10 70	f_1 float64 70 160	f_2 float64 49 100	f_3 float64 6 19	f_4 float64 98 103	f_5 float64 7 90	target int64 0 2
2	30	120	80	6.9	101	76	2
3	40	120	90	12	98	80	0
7	15	76	49	7.9	98	77	1
10	16	100	70	7.5	98	80	1
12	55	100	65	6.1	98	66	1
14	24	120	80	7.5	98	66	1
15	50	120	80	7	98	70	2
16	44	120	90	16	98	80	2
18	28	120	90	7.5	98	70	1
19	35	85	60	11	102	86	0
21	21	75	50	7.8	98	60	1
22	22	100	65	12	98	80	0
23	29	100	70	6.8	98	80	1
24	55	120	90	6.8	98	66	2
27	22	100	65	6.8	98	88	1
28	35	140	100	7.5	98	66	0
30	54	140	100	15	98	66	0
31	50	130	100	16	98	76	0
33	55	140	95	19	98	77	0
35	12	90	60	11	102	60	0
36	23	90	60	7.5	98	76	1
37	22	120	85	7.5	98	88	1
38	17	120	80	6.7	102	76	2
39	55	120	90	12	98	70	2
41	48	140	90	15	98	90	0
42	59	120	80	7.5	98	70	1
43	23	130	70	6.8	98	78	2
44	15	90	49	6	98	77	1
45	19	90	65	11	101	70	0
47	56	120	80	13	98	70	0
48	23	90	60	6.4	98	76	1
49	29	130	70	8	98	78	2
50	32	120	90	6.8	98	70	1
51	40	160	100	19	98	77	0
52	43	130	80	18	98	70	2
54	21	120	80	7	98	77	1
55	15	75	49	7.7	98	77	1
57	12	95	60	6.7	98	77	1
58	15	120	80	7.2	98	70	1
60	21	120	80	6.9	98	88	1
61	65	120	90	6.9	103	76	1
62	23	90	60	7.9	98	76	1
63	10	85	65	6.9	98	70	1
64	20	120	76	7.5	98	70	1
65	35	100	60	15	98	80	0
66	19	120	85	7.9	98	60	1
68	29	130	70	6.1	98	78	2
69	23	90	60	6.7	98	76	1
70	23	120	90	7.8	98	60	2
71	19	120	75	7.5	98	66	1
72	15	76	49	7.8	98	77	1
73	55	100	65	7.8	98	66	1
78	55	140	100	18	98	90	0
80	25	120	80	6.8	98	66	2
83	28	90	60	6.9	98	82	1
84	15	78	49	7.5	98	77	1
85	23	120	90	6.1	98	60	1
87	27	120	90	6.8	102	68	2
88	22	120	85	7.8	98	88	1
92	19	120	85	7	98	60	1
94	29	130	70	6.7	98	78	2
95	25	90	70	6.1	98	80	1
96	35	100	70	7.9	98	60	1
100	15	120	80	7	98	70	1
101	18	100	70	6.7	98	76	1
102	21	100	50	6.8	98	60	1
105	30	120	75	6.8	98	70	2
107	30	120	80	7.8	98	70	1
108	23	90	60	7.01	98	76	1
109	25	120	90	7.5	98	80	1
111	30	120	80	6.8	98	70	2
112	17	90	63	7.5	101	70	1
113	50	140	90	15	98	77	0
114	23	120	90	6.7	98	70	1
116	55	100	70	6.8	101	80	2
118	17	90	63	6.9	101	70	2
119	55	140	80	7.2	101	76	0
120	32	120	90	7.5	98	70	1
121	19	90	70	7.8	98	80	1
122	42	120	90	9	98	70	2
124	28	85	60	9	101	86	2
127	13	90	65	7.9	101	80	2
128	37	120	90	11	98	88	0
129	23	99	60	6.8	98	76	1
130	19	120	60	7	98.4	70	1
131	43	120	90	18	98	70	0
132	55	120	80	6.9	102	80	1
134	23	120	90	7.5	98	60	1
136	15	76	49	6.8	98	77	1
138	15	76	68	7	98	77	1
139	28	115	60	7.5	101	86	2
141	35	100	70	7	98	60	1
142	40	140	100	18	98	90	0
144	29	130	70	7.5	98	78	2
146	35	100	70	7.5	98	66	1
147	23	85	65	7.5	98	70	1
148	34	85	60	11	102	86	0
149	60	90	65	6.8	98	77	2
150	23	130	70	6.9	98	70	2
151	25	140	100	7.01	98	80	0

End of preview. Expand in Data Studio

maternal_health_risk

Part of TabBioMed (TabBioMed: A Large-Scale Benchmark for Biomedical Tabular Learning), a benchmark of 94 curated biomedical tabular datasets. Code, preprocessing and documentation: https://github.com/AI-sandbox/TabBioMed.

Talent Benchmark dataset: maternal_health_risk (task=classification). Rows from official train/val/test files are concatenated; TabBioMed applies five stratified outer folds with three-fold inner CV, rather than the original TALENT split boundaries. Exact duplicate rows were collapsed, and feature patterns with conflicting labels were dropped.

Domain	Task	Samples	Features	Classes
Clinical & Healthcare	classification	381	6	3

Splits. 5 stored outer cross-validation folds (data/fold_<i>/), each with train/val/test files (fold 0: 266/38/77 rows). Every row appears in exactly one outer test fold. HPO in the paper uses inner folds drawn from each outer non-test pool.

Files

data/**.parquet: one row per sample. row_idx is the row index in the original TabBioMed array; f_<i>[_<name>] are the features (in order); target (or target_<j> for multi-target tasks) is the label. Categorical feature indices (0-based, among the features): []. Missing values are stored as nulls.
metadata.json: TabBioMed task metadata (task type, categorical indices, source, description).

from datasets import load_dataset
ds = load_dataset("TabBioMed/maternal_health_risk")

A Croissant (1.1) description with Responsible-AI fields is distributed with the TabBioMed release.

Source and license

Source: Talent Benchmark Dataset - maternal_health_risk

This derived dataset is distributed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Please also cite and respect the terms of the original source.

Citation

@inproceedings{tabbiomed2026,
  title     = {TabBioMed: A Large-Scale Benchmark for Biomedical Tabular Learning},
  author    = {Mateo Bernad{\'o}, Pau and Polapragada, Saivenkata Nagavyjayanthi and Bonet, David and Arbiol Rakuljic, Pol and Mogas Pladevall, Laia and Comajoan Cara, Mar{\c{c}}al and Gonzalez Ferrer, Jesus and Abante, Jordi and Mas Montserrat, Daniel and Haussler, David and Ioannidis, Alexander G.},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS), Evaluations and Datasets Track},
  year      = {2026}
}

Downloads last month: 67