← All datasets

Other

ARCENE Serum Cancer

ARCENE combines mass spectrometry measurements of protein abundance in human serum with synthetic distractor features added for a feature selection challenge. The task is to distinguish cancer samples from normal samples using the publicly labelled data.

Binary classificationLicence: CC BY 4.0OpenML-1458Target: Classopenml ↗

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 · Higher Macro-F1 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
ModelMacro-F1MCCBalanced accuracyROC-AUC
TabDPT †0.83320.67750.83930.8876
TabPFN 30.80930.6410.81470.9164
RealTabPFN 2.50.80660.61940.8090.9257
CatBoost0.79680.60340.80020.8817
AutoGluon0.79380.60580.79790.8773
TabPFN Wide 5k (ne3) ‡0.79180.60260.79420.9108
RealMLP0.77820.58920.78260.8884
RealTabPFN v20.77140.5760.77430.8521
MLP0.76090.55640.76660.8424
KNN0.75970.53040.76250.8719
TabPFN Wide (8k) ‡0.75210.53740.76050.8878
TabM0.74980.51670.75090.8479
Extra Trees0.74180.5020.74610.8532
TabFM0.73920.54610.74130.9037
Random Forest0.73610.4890.74110.8614
MITRA0.73570.49250.73710.8553
Logistic Regression0.71540.45430.71850.8276
XGBoost0.70690.42660.70870.8274
LightGBM0.67690.3650.6770.7898
TabICL v20.358900.50.5