You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
When running weak segments on multi label - in some cases it cruses. problem original is incomplete support of tabular in multi-label, specifically our MyModelWrapper class.
To Reproduce
from deepchecks.nlp.datasets.classification import just_dance_comment_analysis
Describe the bug
When running weak segments on multi label - in some cases it cruses. problem original is incomplete support of tabular in multi-label, specifically our MyModelWrapper class.
To Reproduce
from deepchecks.nlp.datasets.classification import just_dance_comment_analysis
train_ds, test_ds = just_dance_comment_analysis.load_data(include_embeddings=True)
train_probas, test_probas = just_dance_comment_analysis.load_precalculated_predictions(pred_format='probabilities')
from deepchecks.nlp.checks import PropertySegmentsPerformance
check = PropertySegmentsPerformance(segment_minimum_size_ratio = 0.07)
check = check.add_condition_segments_relative_performance_greater_than(0.2)
result = check.run(test_ds, probabilities=test_probas)
result.show()
The text was updated successfully, but these errors were encountered: