MODEL_CARD.md

Model card - tox21_rf_classifier

Model details

• Model name: Random Forest Tox21 Baseline
• Developer: JKU Linz
• Paper URL: https://link.springer.com/article/10.1023/A:1010933404324
• Model type / architecture:
  • Random Forest implemented using sklearn.RandomForestClassifier.
  • Hyperparameters: link to config
  • A separate single-task RF is trained for each Tox21 target.
• Inference: Access via FastAPI. Upon a Tox21 prediction request, a target-specific RF model is called separately for each target; outputs are collected across all single-task models and returned.
• Model version: v0
• Model date: 14.10.2025
• Reproducibility: Code for full training is available and enables retraining from scratch.

Intended use

This model serves as a baseline for evaluating and comparing toxicity prediction methods across the 12 Tox21 pathway assays. It is not intended for clinical decision-making without experimental validation.

Metric

Each Tox21 task is evaluated using the area under the receiver operating characteristic curve (AUC). Overall performance is reported as the mean AUC across all tasks.

Training data

Tox21 training and validation sets.

Evaluation data

Tox21 test set.