quality_comparison/extract_sim.py [90:101]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    c_labels = clustering.labels_
    n_clusters = len(set(c_labels)) - (1 if -1 in c_labels else 0)
    # get floor extents in Y
    # each cluster corresponds to points from 1 floor
    floor_extents = []
    core_sample_y = y_coors[clustering.core_sample_indices_]
    core_sample_labels = c_labels[clustering.core_sample_indices_]
    for i in range(n_clusters):
        floor_min = core_sample_y[core_sample_labels == i].min().item()
        floor_max = core_sample_y[core_sample_labels == i].max().item()
        floor_mean = core_sample_y[core_sample_labels == i].mean().item()
        floor_extents.append({"min": floor_min, "max": floor_max, "mean": floor_mean})
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



scale_comparison/metrics.py [92:102]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    c_labels = clustering.labels_
    n_clusters = len(set(c_labels)) - (1 if -1 in c_labels else 0)
    # estimate floor extents
    floor_extents = []
    core_sample_y = y_coors[clustering.core_sample_indices_]
    core_sample_labels = c_labels[clustering.core_sample_indices_]
    for i in range(n_clusters):
        floor_min = core_sample_y[core_sample_labels == i].min().item()
        floor_max = core_sample_y[core_sample_labels == i].max().item()
        floor_mean = core_sample_y[core_sample_labels == i].mean().item()
        floor_extents.append({"min": floor_min, "max": floor_max, "mean": floor_mean})
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -