kosmos.ml.datasets.qsar_dataset¶
Classes¶
- class QSARDataset(*, min_max_scaler: bool = True)¶
Bases:
kosmos.ml.datasets.dataset.SLDatasetQSAR biodegration dataset for binary classification.
Notes
Number of instances: 1055
Number of features: 41 numeric
Classes: 2 (slightly imbalanced, RD (ready biodegradable) ca 34%; NRB ca 66%)
References
UCI Machine Learning Repository — QSAR dataset: https://archive.ics.uci.edu/dataset/254/qsar+biodegradation
Initialize the dataset.
- Parameters:
min_max_scaler (bool) – Whether to apply min-max scaling to the features.
Properties
- property class_names¶
NRB->0 (not ready biodegradable) RB->1.
- Type:
Return human-readable class labels and map
- property input_dimension¶
Number of feature columns.
- property output_dim¶
Number of distinct classes.