torchdrug.metrics¶
Basic Metrics¶
AUROC¶
-
area_under_roc(pred, target)[source]¶ Area under receiver operating characteristic curve (ROC).
- Parameters
pred (Tensor) – predictions of shape \((n,)\)
target (Tensor) – binary targets of shape \((n,)\)
-
AUROC()¶ alias of
torchdrug.metrics.area_under_roc
AUPRC¶
-
area_under_prc(pred, target)[source]¶ Area under precision-recall curve (PRC).
- Parameters
pred (Tensor) – predictions of shape \((n,)\)
target (Tensor) – binary targets of shape \((n,)\)
-
AUPRC()¶ alias of
torchdrug.metrics.area_under_prc
R2¶
Variadic Accuracy¶
-
variadic_accuracy(input, target, size)[source]¶ Compute classification accuracy over variadic sizes of categories.
Suppose there are \(N\) samples, and the number of categories in all samples is summed to :math`B`.
- Parameters
input (Tensor) – prediction of shape \((B,)\)
target (Tensor) – target of shape \((N,)\). Each target is a relative index in a sample.
size (Tensor) – number of categories of shape \((N,)\)
Chemical Metrics¶
SA¶
-
SA(pred)[source]¶ Synthetic accesibility score.
- Parameters
pred (PackedMolecule) – molecules to evaluate
QED¶
-
QED(pred)[source]¶ Quantitative estimation of drug-likeness.
- Parameters
pred (PackedMolecule) – molecules to evaluate
Chemical Validity¶
-
chemical_validity(pred)[source]¶ Chemical validity of molecules.
- Parameters
pred (PackedMolecule) – molecules to evaluate
LogP¶
-
logP(pred)[source]¶ Logarithm of partition coefficient between octanol and water for a compound.
- Parameters
pred (PackedMolecule) – molecules to evaluate
Penalized LogP¶
-
penalized_logP(pred)[source]¶ Logarithm of partition coefficient, penalized by cycle length and synthetic accessibility.
- Parameters
pred (PackedMolecule) – molecules to evaluate