Never go a step-back in Checkpoints and change the utter_default

DanielZitei · October 11, 2018, 8:11pm

How to never go a step-back in rasa-core conversations? what i have to do in my stories? I have a default answer for every point of my conversation. How i can change the utter_default with progressing my stories?

akelad · October 12, 2018, 3:40pm

What do you mean exactly? If you don’t want to use the fallback policy, then you don’t have to use it

DanielZitei · October 15, 2018, 1:50pm

I train my stories and i can go a step back. the user can go a step foward, but at the same time he can do the step back. Looking the code bellow you gonna see what happend with my application. I think so someone asked this, but i dont the answer.

Logs:

2018-10-15 13:26:16 INFO     root  - Rasa process starting
2018-10-15 13:26:19 INFO     rasa_nlu.components  - Added 'nlp_spacy' to component cache. Key 'nlp_spacy-pt'.
Using TensorFlow backend.
2018-10-15 13:26:19.978026: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2018-10-15 13:26:20 WARNING  py.warnings  - /opt/conda/envs/rasa-env/lib/python3.5/site-packages/pykwalify/core.py:99: UnsafeLoaderWarning:
The default 'Loader' for 'load(stream)' without further arguments can be unsafe.
Use 'load(stream, Loader=ruamel.yaml.Loader)' explicitly if that is OK.
Alternatively include the following in your code:

  import warnings
  warnings.simplefilter('ignore', ruamel.yaml.error.UnsafeLoaderWarning)

In most other cases you should consider using 'safe_load(stream)'
  data = yaml.load(stream)

2018-10-15 13:26:20 INFO     root  - Finished loading agent, starting input channel & server.
Bot loaded. Type a message and press enter:
meu nome é daniel
2018-10-15 13:26:23 DEBUG    rasa_core.tracker_store  - Creating a new tracker for id 'default'.
2018-10-15 13:26:25 WARNING  py.warnings  - /opt/conda/envs/rasa-env/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.
  if diff:

2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Received user message 'meu nome é daniel' with intent '{'confidence': 0.8780217817323307, 'name': 'meu_nome_e'}' and entities '[{'extractor': 'ner_crf', 'value': 'daniel', 'start': 11, 'entity': 'PER', 'confidence': 0.9843487911004908, 'end': 17}]'
2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Logged UserUtterance - tracker now has 3 events
2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Current slot values:
        LOC: None
        setor: None
        apresentar: None
        PER: daniel
2018-10-15 13:26:25 DEBUG    rasa_core.policies.memoization  - Current tracker state [None, None, None, {}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:25 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:25 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:25 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_check_pessoa' with prob 1.00.
2018-10-15 13:26:25 INFO     actions  - <rasa_core.trackers.DialogueStateTracker object at 0x7f24289f9fd0>
2018-10-15 13:26:25 INFO     actions  - <rasa_core.domain.TemplateDomain object at 0x7f242bd2b7f0>
Como posso te ajudar daniel? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica
2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Action 'action_check_pessoa' ended with events '[]'
2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Bot utterance 'BotUttered(text: Como posso te ajudar daniel? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica, data: null)'
2018-10-15 13:26:25 DEBUG    rasa_core.policies.memoization  - Current tracker state [None, None, {}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:25 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:25 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:25 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_listen' with prob 1.00.
2018-10-15 13:26:25 DEBUG    rasa_core.processor  - Action 'action_listen' ended with events '[]'
meu nome é daniel
2018-10-15 13:26:30 DEBUG    rasa_core.tracker_store  - Recreating tracker for id 'default'
2018-10-15 13:26:30 WARNING  py.warnings  - /opt/conda/envs/rasa-env/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.
  if diff:

2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Received user message 'meu nome é daniel' with intent '{'confidence': 0.8780217817323307, 'name': 'meu_nome_e'}' and entities '[{'extractor': 'ner_crf', 'value': 'daniel', 'start': 11, 'entity': 'PER', 'confidence': 0.9843487911004908, 'end': 17}]'
2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Logged UserUtterance - tracker now has 8 events
2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Current slot values:
        LOC: None
        setor: None
        apresentar: None
        PER: daniel
2018-10-15 13:26:30 DEBUG    rasa_core.policies.memoization  - Current tracker state [None, {}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:30 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:30 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:30 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_check_pessoa' with prob 1.00.
2018-10-15 13:26:30 INFO     actions  - <rasa_core.trackers.DialogueStateTracker object at 0x7f24289f9fd0>
2018-10-15 13:26:30 INFO     actions  - <rasa_core.domain.TemplateDomain object at 0x7f242bd2b7f0>
Como posso te ajudar daniel? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica
2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Action 'action_check_pessoa' ended with events '[]'
2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Bot utterance 'BotUttered(text: Como posso te ajudar daniel? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica, data: null)'
2018-10-15 13:26:30 DEBUG    rasa_core.policies.memoization  - Current tracker state [{}, {'entity_PER': 1.0, 'prev_action_listen': 1.0,
'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:30 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:30 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:30 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_listen' with prob 1.00.
2018-10-15 13:26:30 DEBUG    rasa_core.processor  - Action 'action_listen' ended with events '[]'
meu nome é luiz
2018-10-15 13:26:36 DEBUG    rasa_core.tracker_store  - Recreating tracker for id 'default'
2018-10-15 13:26:36 WARNING  py.warnings  - /opt/conda/envs/rasa-env/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.
  if diff:

2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Received user message 'meu nome é luiz' with intent '{'confidence': 0.89971524141042, 'name': 'meu_nome_e'}' and entities '[{'extractor': 'ner_crf', 'value': 'luiz', 'start': 11, 'entity': 'PER', 'confidence': 0.9915673691074862, 'end': 15}]'
2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Logged UserUtterance - tracker now has 13 events
2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Current slot values:
        LOC: None
        setor: None
        apresentar: None
        PER: luiz
2018-10-15 13:26:36 DEBUG    rasa_core.policies.memoization  - Current tracker state [{'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:36 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:36 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:36 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_check_pessoa' with prob 1.00.
2018-10-15 13:26:36 INFO     actions  - <rasa_core.trackers.DialogueStateTracker object at 0x7f24289f9fd0>
2018-10-15 13:26:36 INFO     actions  - <rasa_core.domain.TemplateDomain object at 0x7f242bd2b7f0>
Como posso te ajudar luiz? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica
2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Action 'action_check_pessoa' ended with events '[]'
2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Bot utterance 'BotUttered(text: Como posso te ajudar luiz? Me diz o que você precisa, podemos falar sobre: -atendimento -comercial -curriculo -financeiro -rh -onde fica, data: null)'
2018-10-15 13:26:36 DEBUG    rasa_core.policies.memoization  - Current tracker state [{'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}]
2018-10-15 13:26:36 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:36 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:36 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_listen' with prob 1.00.
2018-10-15 13:26:36 DEBUG    rasa_core.processor  - Action 'action_listen' ended with events '[]'
onde fica
2018-10-15 13:26:54 DEBUG    rasa_core.tracker_store  - Recreating tracker for id 'default'
2018-10-15 13:26:54 WARNING  py.warnings  - /opt/conda/envs/rasa-env/lib/python3.5/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.
  if diff:

2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Received user message 'onde fica' with intent '{'confidence': 0.8919682934714805, 'name': 'servico'}' and entities '[{'extractor': 'ner_crf', 'value': 'onde fica', 'start': 0, 'entity': 'setor', 'confidence': 0.9093690493216995, 'end': 9}]'
2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Logged UserUtterance - tracker now has 18 events
2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Current slot values:
        LOC: None
        setor: onde fica
        apresentar: None
        PER: luiz
2018-10-15 13:26:54 DEBUG    rasa_core.policies.memoization  - Current tracker state [{'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_listen': 1.0, 'intent_servico': 1.0, 'entity_setor': 1.0, 'slot_setor_5': 1.0, 'slot_PER_0': 1.0}]
2018-10-15 13:26:54 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:54 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:54 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_check_servico' with prob 1.00.
Qual cidade você gostaria? -Santos -São Paulo -Itajaí, Joinville, Navegantes...
2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Action 'action_check_servico' ended with events '[]'
2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Bot utterance 'BotUttered(text: Qual cidade você gostaria? -Santos -São Paulo -Itajaí, Joinville, Navegantes..., data: null)'
2018-10-15 13:26:54 DEBUG    rasa_core.policies.memoization  - Current tracker state [{'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'entity_PER': 1.0, 'prev_action_listen': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_check_pessoa': 1.0, 'entity_PER': 1.0, 'slot_PER_0': 1.0, 'intent_meu_nome_e': 1.0}, {'prev_action_listen': 1.0, 'intent_servico': 1.0, 'entity_setor': 1.0, 'slot_setor_5': 1.0, 'slot_PER_0': 1.0}, {'intent_servico': 1.0, 'prev_action_check_servico': 1.0,
'entity_setor': 1.0, 'slot_setor_5': 1.0, 'slot_PER_0': 1.0}]
2018-10-15 13:26:54 DEBUG    rasa_core.policies.memoization  - There is no memorised next action
2018-10-15 13:26:54 DEBUG    rasa_core.policies.ensemble  - Predicted next action using policy_2_KerasPolicy
2018-10-15 13:26:54 DEBUG    rasa_core.policies.ensemble  - Predicted next action 'action_listen' with prob 1.00.
2018-10-15 13:26:54 DEBUG    rasa_core.processor  - Action 'action_listen' ended with events '[]'

DialogueTrain:

from rasa_core.policies import FallbackPolicy, KerasPolicy, MemoizationPolicy
from rasa_core.agent import Agent

fallback = FallbackPolicy(fallback_action_name="action_default_fallback",
                          core_threshold=0.8,
                          nlu_threshold=0.7)

agent = Agent('domain.yml', policies=[MemoizationPolicy(),fallback, KerasPolicy()])
training_data = agent.load_data('stories.md')
agent.train(
    training_data,
    epochs=100)

agent.persist('models/current/dialogue')

myGraphHistory:

akelad · October 16, 2018, 4:42pm

hm i’m not sure what you mean, i don’t see the fallback policy jumping in in your conversation history

DanielZitei · October 16, 2018, 6:51pm

the image is correct. As you can see, the log, however, shows the bot doesnt behave corrently. We are able to type the same intent 3 times and receive the answer in the same amount of times as well. The current problem is that we need the bot to answer only once at each intent, accoreding to the same paths of the image above. If the user types again the same intent, the bot should return a pre-defined answer as its out of scope, like “you have already informed that”. The bot should return as an answer “ofensa”, "servico "or “goodbye”

akelad · October 18, 2018, 1:37pm

You should write stories like this then. Also, are you not using the MemoizationPolicy? Because from the logs it looks like only the KerasPolicy is being used

DanielZitei · October 18, 2018, 5:54pm

As you see above I’m using the memoizationPolicy. My rasa_core version is 10.4. If it doest work, what do i have to do to work?

akelad · October 22, 2018, 12:06pm

You have to see what’s going wrong with your stories. If you can share them, that would help

DanielZitei · October 22, 2018, 9:42pm

Story 1

> check_asked_question

servico{“setor”: “atendimento”}

slot{“setor”: “atendimento”}
action_check_servico

> check_goodbye

Story 2

> check_asked_question

servico{“setor”: “onde fica”}

slot{“setor”: “onde fica”}
action_check_servico

> check_goodbye

Story 3

> check_asked_question

servico{“setor”: “comercial”}

slot{“setor”: “comercial”}
action_check_servico

> check_goodbye

Story 4

> check_asked_question

servico{“setor”: “curriculo”}

slot{“setor”: “curriculo”}
action_check_servico

> check_goodbye

Story 5

> check_asked_question

servico{“setor”: “financeiro”}

slot{“setor”: “financeiro”}
action_check_servico

> check_goodbye

Story 6

> check_asked_question

servico{“setor”: “rh”}

slot{“setor”: “rh”}
action_check_servico

> check_goodbye

Story 7

ofensa

utter_ofensa

> check_asked_question

Story 8

> check_greet

greet

utter_greet

> check_asked_question

Story 9

> check_goodbye

goodbye

utter_goodbye

Story 10

meu_nome_e

action_check_pessoa

> check_asked_question

Story 11

meu_nome_e

action_check_pessoa

> check_goodbye

Story 12

ofensa

utter_ofensa

> check_greet

Story 13

meu_nome_e

action_check_pessoa

> check_ofensa

Story 14

> check_ofensa

ofensa

utter_ofensa

> check_greet

Story 15

> check_ofensa

ofensa

utter_ofensa

> check_asked_question

Story 16

> check_ofensa

ofensa

utter_ofensa

> check_goodbye

Story 17

> check_asked_question

servico{“setor”: “atendimento”}

slot{“setor”: “atendimento”}
action_check_servico

ofensa

utter_ofensa

Story 18

> check_asked_question

servico{“setor”: “onde fica”}

slot{“setor”: “onde fica”}
action_check_servico

ofensa

utter_ofensa

Story 19

> check_asked_question

servico{“setor”: “comercial”}

slot{“setor”: “comercial”}
action_check_servico

ofensa

utter_ofensa

Story 20

> check_asked_question

servico{“setor”: “curriculo”}

slot{“setor”: “curriculo”}
action_check_servico

ofensa

utter_ofensa

Story 21

> check_asked_question

servico{“setor”: “financeiro”}

slot{“setor”: “financeiro”}
action_check_servico

ofensa

utter_ofensa

Story 23

> check_asked_question

servico{“setor”: “atendimento”}

slot{“setor”: “atendimento”}
action_check_servico

> check_greet

Story 23

> check_asked_question

servico{“setor”: “atendimento”}

slot{“setor”: “atendimento”}
action_check_servico

> check_greet

Story 24

> check_asked_question

servico{“setor”: “comercial”}

slot{“setor”: “comercial”}
action_check_servico

> check_greet

Story 25

> check_asked_question

servico{“setor”: “curriculo”}

slot{“setor”: “curriculo”}
action_check_servico

> check_greet

Story 26

> check_asked_question

servico{“setor”: “financeiro”}

slot{“setor”: “financeiro”}
action_check_servico

> check_greet

Story 27

> check_asked_question

servico{“setor”: “rh”}

slot{“setor”: “rh”}
action_check_servico

> check_greet

##Story 29

> check_cidade

cidade{“LOC”: “Santos”}

slot{“LOC”: “Santos”}
action_check_cidade

> check_greet

##Story 30

> check_cidade

cidade{“LOC”: “Santos”}

slot{“LOC”: “Santos”}
action_check_cidade

ofensa

utter_ofensa

##Story 31

> check_cidade

cidade{“LOC”: “Santos”}

slot{“LOC”: “Santos”}
action_check_cidade

> check_goodbye

##Story 32

> check_cidade

cidade{“LOC”: “São Paulo”}

slot{“LOC”: “São Paulo”}
action_check_cidade

> check_greet

##Story 33

> check_cidade

cidade{“LOC”: “São Paulo”}

slot{“LOC”: “São Paulo”}
action_check_cidade

ofensa

utter_ofensa

##Story 34

> check_cidade

cidade{“LOC”: “São Paulo”}

slot{“LOC”: “São Paulo”}
action_check_cidade

> check_goodbye

##Story 35

> check_cidade

cidade{“LOC”: “Itajai”}

slot{“LOC”: “Itajai”}
action_check_cidade

> check_greet

##Story 36

> check_cidade

cidade{“LOC”: “Itajai”}

slot{“LOC”: “Itajai”}
action_check_cidade

ofensa

utter_ofensa

##Story 37

> check_cidade

cidade{“LOC”: “Itajai”}

slot{“LOC”: “Itajai”}
action_check_cidade

> check_goodbye

##Story 38

> check_cidade

cidade{“LOC”: “Joinville”}

slot{“LOC”: “Joinville”}
action_check_cidade

> check_greet

##Story 39

> check_cidade

cidade{“LOC”: “Joinville”}

slot{“LOC”: “Joinville”}
action_check_cidade

ofensa

utter_ofensa

##Story 40

> check_cidade

cidade{“LOC”: “Joinville”}

slot{“LOC”: “Joinville”}
action_check_cidade

> check_goodbye

##Story 41

> check_cidade

cidade{“LOC”: “Navegantes”}

slot{“LOC”: “Navegantes”}
action_check_cidade

> check_greet

##Story 42

> check_cidade

cidade{“LOC”: “Navegantes”}

slot{“LOC”: “Navegantes”}
action_check_cidade

ofensa

utter_ofensa

##Story 43

> check_cidade

cidade{“LOC”: “Navegantes”}

slot{“LOC”: “Navegantes”}
action_check_cidade

> check_goodbye

Story 44

> check_asked_question

servico{“setor”: “onde fica”}

slot{“setor”: “onde fica”}
action_check_servico

ofensa

utter_ofensa

Story 45

> check_asked_question

servico{“setor”: “onde fica”}

slot{“setor”: “onde fica”}
action_check_servico

> check_cidade

akelad · October 25, 2018, 9:26am

Ok the answer is simple, you don’t have the entities recognised present in your Core training data. So the story should be meu_nome_e{"PERSON":"value", ...} – including all entities recognised from your logs

Topic		Replies	Views
Pause story and continue where you left off Rasa Open Source	27	8555	May 28, 2019
Action_default_fallback often predicted as next action after correct action was predicted Rasa Open Source	25	6357	October 6, 2020
Can RASA CORE follow exactly the stories defined in the stories.md file? Getting Started with Rasa rasa-stories	5	408	December 12, 2018
After Fallback policy is started Stories stops working Rasa Open Source	30	4176	July 16, 2019
RASA ALLWAY PREDICT NEXT ACTION IS FALLBACK ACTION Rasa Open Source	5	778	July 17, 2019

Never go a step-back in Checkpoints and change the utter_default

Story 1

Story 2

Story 3

Story 4

Story 5

Story 6

Story 7

Story 8

Story 9

Story 10

Story 11

Story 12

Story 13

Story 14

Story 15

Story 16

Story 17

Story 18

Story 19

Story 20

Story 21

Story 23

Story 23

Story 24

Story 25

Story 26

Story 27

Story 44

Story 45

Related topics