Stop DIET classifications with low confidence

so actually in my examples, i had ‘green daal’ as a food item…so diet must have broken it up and trained on each word or something. i removed that example,and it worked. ‘daal’ as an individual word is not a food item in my db.

Help with rules

Man, I think you can help me with another short problem if you dont mind.

The action_listen after action_record_order is extracting other stuff and trying to fill slots, so how do I stop the extraction for that specific action_listen ?

Ahh what I used is this:

• added another slot called fetch_delivery_home_or_shop
• added a function that gets the slot for delivery_at_home_or_collect_from_shop and fills true/false in the new slot
• added a rule with the new slot as condition.

btw, thanks for the formAction option. I’ll try that out …

Why is my rules contradicting?

I solved it already thanks

Help with rules

I have the stories all made and the confidence isn’t going past .64 something… i have my fallback at .7. Should I reduce the fallback confidence?

Help with rules

I have stories like this:

• ask name
• ask address
• ask delivery_at_home_or_collect_from_shop
• ...10-15 lines of other actions
• deny
• confirm_address
• confirm order

what I want is that the confirm address action at the end should only be predicted after the last deny if the slot delivery_at_home_or_col…

Help with intent and entity extraction

also i tried fallback, but that gives this error. help?

but why are the intents classifying wrong? Because the intent is getting classified randomly, doing any other work on it only gives more errors. Say I said “I want to order” and then I say “please give me one green salad”, it can classify that as deny, payment, whatever with 0.5 - 0.6 confidence

Rasa entities regex

no that’ll be counterproductive since “we rice” doesn’t mean anything… I just wanted to know if I add -

• regex: quantity examples: | -[0-9]

to my nlu, will it work to capture only digits into quantity?

Slot Filling help

Yes, SlotSet worked…made two slots. I figured that whole portion out !

TinyML integration

I’m new to tinyML myself… apparently its an optimized form of ML that can be run on much smaller devices like microprocessors

ValueError(f"Data of type {type(v)} is not allowed only {allowed_types} is accepted for {k}.")

Yes, I rechecked my domain and story files… they had indentation problems… Check for pipes and indents… Still if you get the error, try the rasa nlu convert functions. This is an error on our part, not rasa. Sorry for the panic !

ValueError(f"Data of type {type(v)} is not allowed only {allowed_types} is accepted for {k}.")

please someone help me, i’m freaking out. the problem is with tensorflow input perhaps, but I cant change that

Intent 'nlu_fallback' has only 1 training examples! Minimum is 2, training may fail

Hi, have you found a solution to this? Do I need to set an intent named as “nlu_fallback” ?

How do I connect to rasa api webhook using my virtual machine?

I used ngrok. thank you for your replies, sir

How do I connect to rasa api webhook using my virtual machine?

say I have the ip 10.10.10.10 , then two questions:

how do i curl the rasa server api from the terminal of my local machine? I tried

curl -XPOST -i localhost:5005/webhooks/rest/webhooks ‘{“message”:“hello”}’

but that didn’t work. It returns:

HTTP/1.1 404 Not Found

Introduce yourself!

Hey, I’m new to the forum. I’ve been looking into Rasa for a project, so I’ll probably be pretty frequent. Currently in India. It’s nice to be part of this community. Hopefully, I will learn asap!