Rasa version: 1.10.7
Python version: 3.7
Operating system: Ubuntu 18
Issue:
I am using DIET for intent classification and CRFEntityExtractor for entity extraction in a way shown in a config file - first comes DIET and after it CRFEntityExtractor. The whole training seems to be finished properly, but after testing (rasa test nlu) I get all zero scores (precision, recall, f1 score) for all entities.
However, when I switch their places (first CRFEntityExtractor, then DIET) everything works fine.
Why is this happening? Is it even possible to use CRFEntityExtractor after DIET? Any help would be appreciated.
Content of configuration file (config.yml):
language: en
pipeline:
- name: WhitespaceTokenizer
- name: "CountVectorsFeaturizer"
analyzer: 'word'
- name: "CountVectorsFeaturizer"
analyzer: char_wb
min_ngram: 1
max_ngram: 4
- name: "DIETClassifier"
random_seed: 42
intent_classification: True
entity_recognition: False
epochs: 10
learning_rate: 0.0002
embedding_dimension: 60
number_of_transformer_layers: 1
batch_size: 64
hidden_layer_sizes:
text: [256, 128]
- name: "CRFEntityExtractor"