def test_intersect(project):
am = AnnotationManager(project)
input_annotations = pd.read_csv("examples/valid_raw_data/annotations/intersect.csv")
am.import_annotations(input_annotations)
intersection = AnnotationManager.intersection(
am.annotations[am.annotations["set"].isin(["textgrid", "vtc_rttm"])]
).convert_dtypes()
a = intersection[intersection["set"] == "textgrid"]
b = intersection[intersection["set"] == "vtc_rttm"]
columns = a.columns.tolist()
columns.remove("imported_at")
columns.remove("package_version")
columns.remove("merged_from")
pd.testing.assert_frame_equal(
standardize_dataframe(a, columns),
standardize_dataframe(
pd.read_csv("tests/truth/intersect_a.csv"), columns
).convert_dtypes(),
)
pd.testing.assert_frame_equal(
standardize_dataframe(b, columns),
standardize_dataframe(
pd.read_csv("tests/truth/intersect_b.csv"), columns
).convert_dtypes(),
)
def test_intersect(project):
am = AnnotationManager(project)
input_annotations = pd.read_csv(
'examples/valid_raw_data/raw_annotations/intersect.csv')
am.import_annotations(input_annotations)
am.read()
a, b = am.intersection(am.annotations[am.annotations['set'] == 'textgrid'],
am.annotations[am.annotations['set'] == 'vtc_rttm'])
pd.testing.assert_frame_equal(
a.sort_index(axis=1).sort_values(a.columns.tolist()).reset_index(
drop=True).drop(columns=['imported_at']),
pd.read_csv('tests/truth/intersect_a.csv').sort_index(
axis=1).sort_values(a.columns.tolist()).reset_index(
drop=True).drop(columns=['imported_at']))
pd.testing.assert_frame_equal(
b.sort_index(axis=1).sort_values(b.columns.tolist()).reset_index(
drop=True).drop(columns=['imported_at']),
pd.read_csv('tests/truth/intersect_b.csv').sort_index(
axis=1).sort_values(b.columns.tolist()).reset_index(
drop=True).drop(columns=['imported_at']))
import argparse
parser = argparse.ArgumentParser(
description=
'compute agreement measures for all given annotators over a whole dataset')
parser.add_argument('path', help='path to the dataset')
parser.add_argument('--sets', nargs='+', help='sets to include')
args = parser.parse_args()
speakers = ['CHI', 'OCH', 'FEM', 'MAL']
project = ChildProject(args.path)
am = AnnotationManager(project)
am.read()
intersection = AnnotationManager.intersection(am.annotations, args.sets)
segments = am.get_collapsed_segments(intersection)
segments = segments[segments['speaker_type'].isin(speakers)]
vectors = [
grid_to_vector(
segments_to_grid(segments[segments['set'] == s],
0,
segments['segment_offset'].max(),
100,
'speaker_type',
speakers,
none=False), speakers) for s in args.sets
]