DS1
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
DS1 (custom oracles)
93Total Bugs Detected
42
DS1 27 Taxa, 1949 Sites
-7,107.81Marginal Log-Likelihood
30
DS1
816Detected Defects Count
14
DS1 (test)
-7,108.41MLL
11
DS1 QuickDraw (test)
71.51Reconstruction
10
DS1
90.97Rec
10
DS1
-7,157.99ELBO (nats)
9
DS1 1.0 (test)
0.976R2
9
DS1
0.976R-Squared
9
DS1 (test)
87.84Accuracy
8
DS1
0.002Average Wall-Clock Time (s)
8
DS1
0.15Evaluation Time (min)
7
DS1 27 taxa, 1949 sites 1.0 (test)
-7,032.45Mean Log-Likelihood
6
DS1 1.0 (test)
-2.29Gap (nats)
5
DS1 small numbers (n <= 12)
0Optimality Gap (Popt)
4
DS1
0.005KL Divergence
4
DS1 27 taxa, 1949 sites (ground truth)
-7,108.41Marginal Likelihood (ML)
3