question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

--data option fails.

See original GitHub issue

Rasa version: 1.0.1

Rasa X version (if used & relevant):

Python version: 3.6.7

Operating system (windows, osx, …): Ubuntu 16.04

Issue:

  1. –data option does not work in “rasa train”
  2. err msg misleading: No NLU data given. Please provide NLU data in order to train a Rasa NLU model using the ‘–nlu’ argument.

Per http://rasa.com/docs/rasa/user-guide/command-line-interface/

If the ./data directory is present, “rasa train” builds by grabbing data from ./data, however, if ./data is removed, the build fails and displays the err msg despite the explicit use of --data option

And there is no such an option ‘–nlu’ in 1.0.

Error (including full traceback):

No NLU data given. Please provide NLU data in order to train a Rasa NLU model using the '--nlu' argument.

Command or request that led to error:

#!/bin/bash

rasa train -v \
    --config ./nlu_config.yml \
    --data ./NLU_train_data \
    --out ./models \
    nlu

Content of configuration file (config.yml) (if relevant):

language: "en"

pipeline: "supervised_embedding"

Content of domain file (domain.yml) (if relevant):


Issue Analytics

  • State:closed
  • Created 4 years ago
  • Comments:6 (2 by maintainers)

github_iconTop GitHub Comments

1reaction
Lambikcommented, Aug 28, 2019

Oh, through digging through the source code… I got it working, so this is a post for future google searchers:

There is indeed an --nlu option, that is not in the manual. The other thing that is important, is that my model filename has to end with .md

The final train command is: rasa train nlu --nlu /mnt/hgfs/backend/storage/app/rasa/nlu_test.md --out /mnt/hgfs/backend/storage/app/rasa --fixed-model-name nlu_test_model

The final model name will automatically get a .tar.gz extension.

So this issue can remain closed 😃 but maybe docs could be updated a bit.

0reactions
Lambikcommented, Aug 28, 2019

I have this same issue (rasa version 1.2.5).

My command line is this: rasa train --data /mnt/hgfs/backend/storage/app/rasa/nlu_test --out /mnt/hgfs/backend/storage/app/rasa --fixed-model-name nlu_test_model nlu

My nlu file (“nlu_test”) is simple:

## intent:16
- yes
- indeed
- of course
- that sounds good
- correct

## intent:7
- no
- never
- I don't think so
- don't like that
- no way
- not really

## intent:1
- bye
- goodbye
- see you around
- see you later

## intent:22
- hey
- hello
- hi
- good morning
- good evening
- hey there

## intent:30
- perfect
- very good
- great
- amazing
- wonderful
- I am feeling very good
- I am great
- I'm good

## intent:39
- sad
- very sad
- unhappy
- bad
- very bad
- awful
- terrible
- not very good
- extremely sad
- so sad

And I get the error No NLU data given. Please provide NLU data in order to train a Rasa NLU model using the '--nlu' argument.

There is no such thing as an --nlu parameter.

@akelad dunno if I should tag you…

Read more comments on GitHub >

github_iconTop Results From Across the Web

"Data Option Error" on the IDP Smart-70 Printer | ID Wholesaler
“Data Option Error” on the IDP Smart-70 Printer ... Possible Reasons for Error Message Reason #1. The Smart-70's flipper module was removed from...
Read more >
Fails-to-Deliver Data - SEC.gov
File Format Size File: November 2022, first half Format:ZIP Size:1.11 MB File: October 2022, second half Format:ZIP Size:1.35 MB File: October 2022, first half Format:ZIP Size:1.14...
Read more >
Bootstrap Tooltip data[option] is not a function - Stack Overflow
The tooltip shows up fine, the the destroy line throws an error data[option] is not a function . If I change it to...
Read more >
More on data validation - Microsoft Support
To turn off manual recalculation, go to the Formulas tab > Calculation group > Calculation Options > click Automatic. Formulas are error free...
Read more >
Handling operation errors - Apollo GraphQL Docs
Apollo Client ignores partial data by default, but you can override this behavior by setting a GraphQL error policy. Network errors. These are...
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found