Optimize RelationClassifier by adding the option to filter long sentences and truncate context #2571
Annotations
4 errors and 1 warning
|
Run tests:
tests/model_test_utils.py#L109
TestRelationExtractor.test_train_load_use_model[False]
AssertionError: assert 'place_of_birth' == 'founded_by'
- founded_by
+ place_of_birth
|
|
Run tests:
tests/test_datasets.py#L379
test_load_conllu_plus_corpus[False]
assert 5 == 4
+ where 5 = len(<torch.utils.data.dataset.ConcatDataset object at 0x7fe3bbd12490>)
+ where <torch.utils.data.dataset.ConcatDataset object at 0x7fe3bbd12490> = <flair.datasets.sequence_labeling.ColumnCorpus object at 0x7fe418217d00>.train
|
|
Run tests:
tests/test_datasets.py#L396
test_load_conllu_corpus_plus_in_memory[False]
assert 5 == 4
+ where 5 = len(<torch.utils.data.dataset.ConcatDataset object at 0x7fe3bbc33b80>)
+ where <torch.utils.data.dataset.ConcatDataset object at 0x7fe3bbc33b80> = <flair.datasets.sequence_labeling.ColumnCorpus object at 0x7fe3bbc33d00>.train
|
|
Run tests
Process completed with exit code 1.
|
|
ubuntu-latest pipelines will use ubuntu-24.04 soon. For more details, see https://github.com/actions/runner-images/issues/10636
|
Loading