rasa | automate text- and voice-based conversations | Chat library

This is a standard linter error letting you know that you've declared a variable without using it -- sometimes it indicates a typo in your code, or an opportunity to clean up dead code.

If you see this error and you think you are using the variable, look very carefully at the line of code where you're using it, and you'll probably find a subtle problem around scoping, or a typo in the variable name, or something like that.

In this case, the typo is that your f-string is missing its f.

The output does not show any actual errors, so assuming the issue is a time-out during installation or it takes unusually long to install a package, this would be caused by the newly introduced pip dependency resolver. This has been discussed before here. Try to upgrade pip to the newest version, where this issue is reported to be fixed, via

The problem was easy to solve, basically the prediction would have to take into account the previous state of the conversation. Here is a code example:

Upgrade pip:

This error could be due to the Rasa version you're using (output of rasa --version). In the current versions (>2.1), HFTransformersNLP and LanguageModelTokenizer are deprecated. Using a BERT model can be achieved with any tokenizer and

As far as I know, the only WhatsApp alternative is WhatsApp business, which is likely more expensive as you need to have a business license with them. Unfortunately their API is not public like most messaging services are.

These are some of the main differences between Dialog Flow and Rasa:

Dialog Flow:

• No installation needed, you can start using it immediately.

• Easy to use, non-tech people can build a bot.

• Closed system.

• Web-based interface for building bots.

• Data is hosted on the cloud.

• Can’t be hosted on your server or on-premise.

• Out-of-box integration with Google Assistant, Skype, Slack, Facebook Messenger, etc.

Rasa:

• Requires installation of multiple components.

• Requires tech knowledge.

• Open-source code available in GitHub.

• No interface provided, write JSON or markdown files.

• No hosting provided.

• You can host it on your server.

• No out-of-box integration.

Try adding this before installing Rasa X: brew install libxml2 libxmlsec1 pkg-config

There is a dedicated section in the Rasa Docs about reaching out to the user which, I believe, you haven't consulted yet (please, correct me if I'm wrong). Take a look and, if you encounter any issues, please, post on the Rasa forum.

Regarding your setup and why it doesn't achieve what you want: Rasa stories describe what actions a bot should take in reaction to a user message and, if applicable, to previous conversation history. Rasa models "learn" from these stories how to react in context. When a story starts with a bot action, a model cannot learn from such a story because it's unclear what the context is -- after which user message should that action be predicted? And if you're asking "How come it's not clear? The story says that my action should start the conversation!", then it's important to realise that a story doesn't necessarily describe a conversation from the beginning -- it can also represent just a part of a conversation, maybe just the ending. Again, if you have any follow-up questions, look at our docs and ask on the forum :-)