kosmos.ml.datasets.bloodmnist_dataset

Classes

class BloodMNISTDataset(*, min_max_scaler: bool = True)

Bases: kosmos.ml.datasets.dataset.SLDataset

BloodMNIST dataset for blood cell classification from biomedical images.

Notes

  • Number of instances: 17,092 (11,959 train + 1,712 val + 3,421 test)

  • Number of features: 2,352 numeric (28x28x3 RGB images, flattened)

  • Classes: 8 (different blood cell types)

References

Initialize the dataset.

Parameters:

min_max_scaler (bool) – Whether to apply min-max scaling to the features.


Properties

property class_names

Return human-readable class labels for blood cell types.

property input_dimension

Number of feature columns.

property output_dim

Number of distinct classes.