I used the below config but as i train the model an error is thrown
language : “en”
pipeline:
-
name: “tokenizer_whitespace”
-
name: “ner_crf”
-
name: “intent_featurizer_count_vectors”
“token_pattern”: r’(?u)\b\w+\b’
-
name: “intent_classifier_tensorflow_embedding”
am i doing something wrong ?