@netcaver they have custom built on haptik. and have very limited functions defined in there data file in form of csv files.
Can’t we do it by using oov token of CountVectorsFeaturizer? I tried the following but didn’t work.
## intent:inform - My name is [James](name) - my name is [Leota](name) - My name is oov - oov - [Louis](name)
language: en pipeline: - name: SpacyNLP case_sensitive: False - name: SpacyTokenizer - name: SpacyFeaturizer - name: RegexFeaturizer - name: LexicalSyntacticFeaturizer - name: CountVectorsFeaturizer OOV_token: oov token_pattern: (?u)\b\w+\b - name: CountVectorsFeaturizer analyzer: char_wb min_ngram: 1 max_ngram: 4 - name: DIETClassifier epochs: 100 - name: EntitySynonymMapper policies: - name: MemoizationPolicy - name: TEDPolicy max_history: 5 epochs: 100 - name: MappingPolicy - name: FormPolicy
Another option is asking the first name and last name and accepting user input only if it includes two words. if not, use a new utterance as “please provide one first name and one last name”.