Density Estimation Benchmark Datasets

A collection of datasets used in machine learning for density estimation.

If you use any of the datasets, you should cite their original papers.<sup id="a1">1</sup><sup id="a1">2</sup><sup id="a1">3</sup><sup id="a1">4</sup>

Datasets

|Dataset | type | #vars | #train | #valid | #test | density | abbrv | |:------|:---:|---:|---:|---:|---:|---:|---:| |NLTCS<sup id="a1">1</sup>| binary | 16 | 16181 | 2157 | 3236 | 0.332|NLTCS| |MSNBC<sup id="a1">1</sup>| binary | 17 | 291326 | 38843 | 58265 | 0.166|msnbc| |KDDCup2k<sup id="a1">1</sup>| binary | 65 | 180092 | 19907 | 34955 | 0.008|kdd| |Plants<sup id="a1">1</sup>| binary | 69 | 17412 | 2321 | 3482 | 0.180|plants| |Audio<sup id="a1">1</sup>| binary | 100 | 15000 | 2000 | 3000 | 0.199|baudio| |Jester<sup id="a1">1</sup>| binary | 100 | 9000 | 1000 | 4116 | 0.608|jester| |Netflix<sup id="a1">1</sup>| binary | 100 | 15000 | 2000 | 3000 | 0.541|bnetflix| |Accidents<sup id="a1">2</sup>| binary | 111 | 12758 | 1700 | 2551 | 0.291|accidents| |Mushrooms<sup id="a3">4</sup>| binary | 112 | 2000 | 500 | 5624 | 0.187|mushrooms| |Adult<sup id="a3">4</sup>| binary | 123 | 5000 | 1414 | 26147 | 0.112|adult| |Connect 4<sup id="a3">4</sup>| binary | 126 | 16000 | 4000 | 47557 | 0.333|connect4| |OCR Letters<sup id="a3">4</sup>| binary | 128 | 32152 | 10000 | 10000 | 0.220|ocr_letters| |RCV-1<sup id="a3">4</sup>| binary | 150 | 40000 | 10000 | 150000 | 0.138|rcv1| |Retail<sup id="a2">2</sup>| binary | 135 | 22041 | 2938 | 4408 | 0.024|tretail| |Pumsb-star<sup id="a2">2</sup>| binary | 163 | 12262 | 1635 | 2452 | 0.270|pumsb_star| |DNA<sup id="a2">2</sup>| binary | 180 | 1600 | 400 | 1186 | 0.253|dna| |Kosarek<sup id="a2">2</sup>| binary | 190 | 33375 | 4450 | 6675 | 0.020|kosarek| |MSWeb<sup id="a1">1</sup>| binary | 294 | 29441 | 3270 | 5000 | 0.010|MSWeb| |NIPS<sup id="a3">4</sup>| binary | 500 | 400 | 100 | 1240 | 0.367|nips| |Book<sup id="a1">1</sup>| binary | 500 | 8700 | 1159 | 1739 | 0.016|book| |EachMovie<sup id="a1">1</sup>| binary | 500 | 4525 | 1002 | 591 | 0.059|tmovie| |WebKB<sup id="a1">1</sup>| binary | 839 | 2803 | 558 | 838 | 0.064|cwebkb| |Reuters-52<sup id="a1">1</sup>| binary | 889 | 6532 | 1028 | 1540 | 0.036|cr52| |20 NewsGroup<sup id="a1">1</sup>| binary | 910 | 11293 | 3764 | 3764 | 0.049|c20ng| |Movie reviews<sup id="a1">3</sup>| binary | 1001 | 1600 | 150 | 250 | 0.140 |moviereview| |BBC<sup id="a2">2</sup>| binary | 1058 | 1670 | 225 | 330 | 0.078|bbc| |Voting<sup id="a1">3</sup>| binary | 1359 | 1214 | 200| 350| 0.333|voting| |Ad<sup id="a2">2</sup>| binary | 1556 | 2461 | 327 | 491 | 0.008|ad|

Introduced in:

<b id="f1">1</b> Daniel Lowd, Jesse Davis: Learning Markov Network Structure with Decision Trees. ICDM 2010

<b id="f2">2</b> Jan Van Haaren, Jesse Davis: Markov Network Structure Learning: A Randomized Feature Generation Approach. AAAI 2012

<b id="f3">3</b> Jessa Bekker, Jesse Davis, Arthur Choi, Adnan Darwiche, Guy Van den Broeck: Tractable Learning for Complex Probability Queries. NIPS 2015

<b id="f4">4</b> Hugo Larochelle, Iain Murray: The Neural Autoregressive Distribution Estimator. AISTATS 2011