Other

mtp2

The mtp2 dataset represents drug-like molecules with numerical molecular descriptors calculated using Adriana.Code. Its regression target, stored as oz1143, is the compound's melting point, drawn from a study of molecular structure and the melting behavior of solid drugs.

RegressionLicence: PublicOpenML-430Target: oz1143openml ↗

At this operating point

Model comparison

p=10,000, n=100 · Lower RMSE is better

Random Forest is marked with a diamond. Models are ordered by the selected metric.

Across sample budgets

Learning curves

Models

All metric values at this operating point

Model	RMSE	MAE	R²
TabPFN 3.5	0.146	0.1152	0.3291
TabFM	0.149	0.1188	0.3006
RealTabPFN v2	0.1503	0.1178	0.2902
RealTabPFN 2.5	0.1511	0.1185	0.2819
Kumo Tabular (M)	0.1511	0.1181	0.2743
LimiX2	0.1521	0.1192	0.2706
Extra Trees (tuned)	0.1538	0.1198	0.2516
Extra Trees	0.1543	0.1194	0.2525
AutoGluon	0.1544	0.1229	0.2443
Random Forest (tuned)	0.1551	0.1213	0.2444
Random Forest	0.1569	0.1219	0.226
MITRA	0.1574	0.1219	0.2207
TabM	0.1583	0.1219	0.2137
Causilo	0.1592	0.1223	0.2047
XGBoost (tuned)	0.1612	0.1285	0.1734
MLP (tuned)	0.1626	0.1267	0.1607
GeneICL	0.1657	0.1293	0.1391
MLP	0.1662	0.1282	0.1287
TabDPT	0.1665	0.1277	0.1191
XGBoost	0.1711	0.1326	0.08129
Logistic Regression (tuned)	0.1735	0.135	0.04981
RealMLP	0.1801	0.1469	-0.05221
Logistic Regression	0.1879	0.1496	-0.1209