rasa.core.processor - [debug ] processor.message.parse parse_data_entities= parse_data_intent={‘name’: ‘health_status’, ‘confidence’: 0.9846216440200806} parse_data_text=美国总统特朗普 config： language: zh #pipeline: null pipeline: - name: JiebaTokenizer - name: CountVectorsFeaturizer token_pattern: "(?u)\\b\\w+\\b" # 中文分词的正则表达式 min_ngram: 1 max_ngram: 2 - name: CountVectorsFeaturizer analyzer: char_wb min_ngram: 1 max_ngram: 3 - name: LexicalSyntacticFeaturizer - name: LanguageModelFeaturizer model_name: "bert" model_weights: "bert-base-chinese" - name: DIETClassifier epochs: 100 learning_rate: 0.002 number_of_transformer_layers: 2 constrain_similarities: true intent_classification_threshold: 0.8 - name: FallbackClassifier

Intention recognition error

haoyy (haoyy) August 5, 2024, 9:10am 2

Please help take a look and see where the problem is.

Topic		Replies	Views
2.0.0rc1， Train successfully but not work! Rasa Open Source	1	675	September 29, 2020
Rasa_nlu returns intent as null for exact copy of text given during training Rasa Open Source	2	908	October 10, 2018
Intent incorrectly recognised with high confidence Rasa Open Source	3	344	June 20, 2023
Rasa classifies random input as intents with high probability Rasa Open Source	24	1789	April 20, 2023
Failing at intent classification Rasa Open Source	4	854	August 5, 2019