Can not get entities by using SpacyNLP

Rasa Open Source
1

I just test Rasa2 with SpacyNLP3, can not get the entity in response .

Codes in utl/rasa/spacy at main · utlai/utl · GitHub.

config.yaml

language: zh
pipeline:
  - name: SpacyNLP                   # 预训练词向量
    model: zh_core_web_trf
  - name: SpacyTokenizer          # 文本分词器
  - name: SpacyEntityExtractor  #文本特征化
  - name: SpacyFeaturizer         #特征提取器 将一句话变成一个向量
    pooling: mean
  - name: CountVectorsFeaturizer #创建用户信息和标签(意图和响应)的词袋表征 为意图分类和 response selection创建特征
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier              #意图分类
    epochs: 100
  - name: EntitySynonymMapper   #同义实体
  - name: RegexFeaturizer
  - name: ResponseSelector
    epochs: 100

policies:
  - name: MemoizationPolicy
  - name: TEDPolicy
    max_history: 5
    epochs: 100
  - name: MappingPolicy

domain.yaml

 version: "2.0"

intents:
  - opt_log
  - opt_test

entities:
  - print
  - loglevel
  - message

slots:
  print:
     type: text
     influence_conversation: false
  loglevel:
    type: text
    influence_conversation: false
  message:
    type: text
    influence_conversation: false

responses:
  utter_greate:
  - text: "fine"

session_config:
  session_expiration_time: 60
  carry_over_slots_to_new_session: true

nlu.yml

version: "2.0"
nlu:
- intent: opt_log
  examples: |
    - [打印]{"entity":"print", "value":"syn_print"}[消息]{"entity":"loglevel", "value":"lkp_loglevel"}[内容](message)
    - [打印]{"entity":"print", "value":"syn_print"}[消息]{"entity":"loglevel", "value":"lkp_loglevel"}日志[内容](message)

I send below request to nlu server.

{
    "text": "打印错误测试"
}

Why the return msg has no lookup named “loglevel” and entity named “message”? Just has synonym “print” value in it which are right.

{
    "text": "打印错误测试",
    "intent": {
        "id": 6128636546715775035,
        "name": "opt_log",
        "confidence": 0.9967185854911804
    },
    "entities": [
        {
            "entity": "print",
            "start": 0,
            "end": 2,
            "confidence_entity": 0.9995416402816772,
            "value": "syn_print",
            "extractor": "DIETClassifier",
            "processors": [
                "EntitySynonymMapper"
            ]
        }
    ],
    "intent_ranking": [
        {
            "id": 6128636546715775035,
            "name": "opt_log",
            "confidence": 0.9967185854911804
        },
        {
            "id": -1888143233846630085,
            "name": "opt_test",
            "confidence": 0.0032813996076583862
        }
    ],
    "response_selector": {
        "all_retrieval_intents": [],
        "default": {
            "response": {
                "id": null,
                "responses": null,
                "response_templates": null,
                "confidence": 0.0,
                "intent_response_key": null,
                "utter_action": "utter_None",
                "template_name": "utter_None"
            },
            "ranking": []
        }
    }
}

By the way,

1. How to remove below warning?

huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
        - Avoid using `tokenizers` before the fork if possible
        - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)

2. How to remove below Warning?

/usr/local/lib/python3.8/site-packages/rasa/utils/train_utils.py:558: UserWarning: constrain_similarities is set to `False`. It is recommended to set it to `True` when using cross-entropy loss. It will be set to `True` by default, Rasa Open Source 3.0.0 onwards.
  rasa.shared.utils.io.raise_warning(
/usr/local/lib/python3.8/site-packages/rasa/utils/train_utils.py:531: UserWarning: model_confidence is set to `softmax`. It is recommended to try using `model_confidence=linear_norm` to make it easier to tune fallback thresholds.
  rasa.shared.utils.io.raise_warning(