diff options
| author | John Bauer <horatio@gmail.com> | 2022-11-06 04:27:34 +0300 |
|---|---|---|
| committer | John Bauer <horatio@gmail.com> | 2022-11-06 04:27:34 +0300 |
| commit | 96a43d2c9bcb04989eb27a6d96dd5ef5d8c9dc4d (patch) | |
| tree | dd2151d4fab271675ada4da0a67da4cb8d94bb35 | |
| parent | 35637568473df11de95d0a6e35d0b41c1ea10e48 (diff) | |
Add the ability to quiet the logging
| -rw-r--r-- | stanza/models/common/bert_embedding.py | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/stanza/models/common/bert_embedding.py b/stanza/models/common/bert_embedding.py index 627076df..65a7fb16 100644 --- a/stanza/models/common/bert_embedding.py +++ b/stanza/models/common/bert_embedding.py @@ -65,7 +65,7 @@ def tokenize_manual(model_name, sent, tokenizer): return tokenized, tokenized_sent -def filter_data(model_name, data, tokenizer = None): +def filter_data(model_name, data, tokenizer = None, log_level=logging.INFO): """ Filter out the (NER) data that is too long for BERT model. """ @@ -82,7 +82,7 @@ def filter_data(model_name, data, tokenizer = None): filtered_data.append(sent) - logger.info("Eliminated %d of %d datapoints because their length is over maximum size of BERT model. ", (len(data)-len(filtered_data)), len(data)) + logger.log(log_level, "Eliminated %d of %d datapoints because their length is over maximum size of BERT model.", (len(data)-len(filtered_data)), len(data)) return filtered_data |