← All datasets

Molecular properties

Lipophilicity (AstraZeneca)

AstraZeneca's lipophilicity dataset measures how compounds distribute between octanol and water at pH 7.4. The benchmark uses molecular fingerprints to predict logD, the logarithm of this distribution coefficient.

RegressionLicence: CC BY 4.0 via TDC; underlying source terms require verification before re-hostingTDC-Lipophilicity-AstraZenecaTarget: Ytdc ↗

Mean cross-validation scores for this dataset. Each score comes from the requested feature- and sample-cap cell, and failed fits use the constant predictor.

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

All metric values at this operating point
ModelRMSEMAER²
Extra Trees1.1750.93380.04187
Random Forest1.1760.93450.04061
RealTabPFN 2.51.1810.94930.03327
CatBoost1.1820.9520.03173
TabM1.1850.95240.02718
RealTabPFN v21.1860.95610.02603
XGBoost1.1860.950.02504
TabPFN 31.190.94620.01797
TabDPT †1.1910.94590.01539
MITRA1.1990.97060.003576
MLP1.2090.9691-0.01252
LightGBM1.2110.9774-0.01933
Logistic Regression1.2130.9655-0.02234
TabICL v21.2260.9649-0.0462
AutoGluon1.2270.9649-0.04633
TabFM1.2590.9887-0.1031
RealMLP1.2681.005-0.1137
KNN1.281.051-0.1362