Data lost, so not saved from the Bot Response page in multi-language

[scilaci]

Hi,

I'm using this amazing tool but I had a problem and I lost 2 times my data in "Bot Responses"

I'm using the multi-language and I was working in first with the french data.
when It was done I add the dutch version of data (in Bot Responses).

After completing these missing data in this page:

I trained the bot.
After that I selected the dutch language and I writed some sentences in the chatbot but It was not working in dutch... no answers.

In the "Incoming" page, nothing new inside these 2 tab "New utterances" and "Conversations"

So the Botfront was looking blocked and I decided to restart with "docker-compose restart"
but I lost everything in the Bot Responses page (in dutch)

I find a workaround, I need for that to type "the bot responses text" from the stories section
And from there the data are saved and I can restart the BotFront

OS: Windows 10

$ docker-compose version
docker-compose version 1.26.2, build eefe0d31
docker-py version: 4.2.2
CPython version: 3.7.4
OpenSSL version: OpenSSL 1.1.1c 28 May 2019

$ docker version
Client: Docker Engine - Community
Version: 19.03.12
API version: 1.40
Go version: go1.13.10
Git commit: 48a66213fe
Built: Mon Jun 22 15:43:18 2020
OS/Arch: windows/amd64
Experimental: false

$ botfront --version
0.25.9

Log after trainning the bot (data from Bot Response page ):

botfront-app | 2020-08-10T15:23:08.854Z [info] : POST at /data/convert/ user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.getTrainingPayload with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"},"language":""} url: http://rasa:5005/data/convert/ botfront-rasa | 2020-08-10 15:23:08 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpbqm0x01p/nlu_data' is 'rasa_nlu'. botfront-app | 2020-08-10T15:23:08.873Z [info] : POST at http://rasa:5005/data/convert/ succeeded user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.getTrainingPayload with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"},"language":""} 200 url: http://rasa:5005/data/convert/ botfront-app | 2020-08-10T15:23:08.875Z [info] : POST at /data/convert/ user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.getTrainingPayload with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"},"language":""} url: http://rasa:5005/data/convert/ botfront-rasa | 2020-08-10 15:23:08 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpj7b8n7yo/nlu_data' is 'rasa_nlu'. botfront-app | 2020-08-10T15:23:08.888Z [info] : POST at http://rasa:5005/data/convert/ succeeded user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.getTrainingPayload with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"},"language":""} 200 url: http://rasa:5005/data/convert/ botfront-app | 2020-08-10T15:23:08.890Z [info] : POST at /model/train user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.train with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"}} url: http://rasa:5005/model/train botfront-rasa | 2020-08-10 15:23:08 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/stories.md' is 'unk'. botfront-rasa | 2020-08-10 15:23:08 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/fr.md' is 'md'. botfront-rasa | 2020-08-10 15:23:08 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/nl.md' is 'md'. botfront-rasa | /opt/venv/lib/python3.7/site-packages/rasa/core/domain.py:151: FutureWarning: No tracker session configuration was found in the loaded domain. Domains without a session config will automatically receive a session expiration time of 60 minutes in Rasa version 2.0 if not configured otherwise. botfront-rasa | session_config = cls._get_session_config(data.get(SESSION_CONFIG_KEY, {})) botfront-rasa | 2020-08-10 15:23:09 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/fr.md' is 'md'. botfront-rasa | 2020-08-10 15:23:09 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/nl.md' is 'md'. botfront-rasa | 2020-08-10 15:23:09 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/fr.md' is 'md'. botfront-rasa | 2020-08-10 15:23:09 DEBUG rasa.nlu.training_data.loading - Training data format of '/tmp/tmpli9v3zh2/nlu/nl.md' is 'md'. botfront-rasa | 2020-08-10 15:23:09 DEBUG rasa.model - Extracted model to '/tmp/tmpdyz8tns3'. botfront-app | 2020-08-10T15:23:10.011Z [info] : POST at http://rasa:5005/model/train succeeded user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.train with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"}} 200 url: http://rasa:5005/model/train botfront-app | 2020-08-10T15:23:10.353Z [info] : PUT at /model user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.train with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"}} url: http://rasa:5005/model botfront-rasa | 2020-08-10 15:23:10 DEBUG rasa_addons.core.tracker_stores.botfront - BotfrontTrackerStore tracker store created botfront-rasa | 2020-08-10 15:23:10 DEBUG rasa.core.tracker_store - Connected to BotfrontTrackerStore. botfront-rasa | 2020-08-10 15:23:11 DEBUG rasa.model - Extracted model to '/tmp/tmp_5myewts'. botfront-rasa | 2020-08-10 15:23:11 DEBUG rasa.utils.tensorflow.models - Loading the model ... botfront-rasa | 2020-08-10 15:23:11 DEBUG rasa.utils.tensorflow.models - Finished loading the model. botfront-rasa | 2020-08-10 15:23:11 DEBUG rasa.utils.tensorflow.models - Building tensorflow prediction graph... botfront-rasa | 2020-08-10 15:23:12 DEBUG rasa.utils.tensorflow.models - Finished building tensorflow prediction graph. botfront-rasa | 2020-08-10 15:23:12 DEBUG rasa.utils.tensorflow.models - Loading the model ... botfront-rasa | 2020-08-10 15:23:12 DEBUG rasa.utils.tensorflow.models - Finished loading the model. botfront-rasa | 2020-08-10 15:23:12 DEBUG rasa.utils.tensorflow.models - Building tensorflow prediction graph... botfront-rasa | 2020-08-10 15:23:13 DEBUG rasa.utils.tensorflow.models - Finished building tensorflow prediction graph. botfront-rasa | 2020-08-10 15:23:13 DEBUG rasa.utils.tensorflow.models - Loading the model ... botfront-rasa | 2020-08-10 15:23:14 DEBUG rasa.utils.tensorflow.models - Finished loading the model. botfront-rasa | 2020-08-10 15:23:14 DEBUG rasa.utils.tensorflow.models - Building tensorflow prediction graph... botfront-rasa | 2020-08-10 15:23:14 DEBUG rasa.utils.tensorflow.models - Finished building tensorflow prediction graph. botfront-rasa | 2020-08-10 15:23:14 INFO rasa.core.policies.ensemble - MappingPolicy not included in policy ensemble. Default intents 'restart and back will not trigger actions 'action_restart' and 'action_back'. botfront-rasa | 2020-08-10 15:23:14 DEBUG rasa.core.nlg.generator - Instantiated NLG to 'GraphQLNaturalLanguageGenerator'. botfront-rasa | 2020-08-10 15:23:14 DEBUG rasa.server - Successfully loaded model '/app/models/model-bf.tar.gz'. botfront-app | 2020-08-10T15:23:14.611Z [info] : PUT at http://rasa:5005/model succeeded user: bNhRsAXDpynTxPbHm /imports/api/instances/instances.methods.js - rasa.train with {"projectId":"bf","instance":{"_id":"KKgsaYLQAhEZEy8bz","name":"Default Instance","host":"http://rasa:5005","projectId":"bf"}} 204 url: http://rasa:5005/model

[znat]

Hi @scilaci , sorry about that!
I was trying to reproduce I am not sure I understand the steps.

Are you creating your responses from the responses screen?
You create one language, click the Dutch tab, enter response, and then it blocks?
What happen when it blocks?
Any chance you can upload a screencast (e.g. with https://www.screencastify.com/)
Thanks a lot!

[scilaci]

Hi Thanks for your reply,

Are you creating your responses from the responses screen?
Yes

You create one language, click the Dutch tab, enter response, and then it blocks?
Yes like that,
when I finished to create the french part, I decided to add the dutch language and in the response screen I completed all empty response.

What happen when it blocks?_

In fench:
in the chatbot everythings is working like before, response ok, stories ok

In dutch:
Is looking like not training the bot with last change or not using the last trained file, because the chatbot is not responding (not understanding my intent).
All what I'm writing in the chatbot is not reported In the "Incoming" page, nothing new inside these 2 tab "New utterances" and "Conversations"

I can reproduce It easelly:
Juste adding something in the response page and restarting the botfront
after that I lose my last change

Maybe is for that, that the bot is not trained correctly ?