Hosting validator models doc #1008
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dtam
reviewed
Aug 13, 2024
| @@ -0,0 +1,210 @@ | |||
| # Hosting Validator Models | |||
|
|
|||
| Validation using machine learning models is a highly effective way of detecting LLM hallucinations that are not easily or possibly detected using traditional coding techniques. We’ve selected and tuned some ML models to run different validations. These models are wrapped within model validators. This makes the access pattern straightforward — you can use model-based validators the same as any other [validator](https://www.guardrailsai.com/docs/concepts/validators). | |||
There was a problem hiding this comment.
nit links to https://www.guardrailsai.com/docs should be relative and not hard coded to this domain so local and staging can mostly work
dtam
reviewed
Aug 13, 2024
zsimjee
reviewed
Aug 14, 2024
zsimjee
reviewed
Aug 14, 2024
zsimjee
reviewed
Aug 14, 2024
|
|
||
| The standard inference request can be found [here](https://github.com/guardrails-ai/guardrails/blob/main/guardrails/validator_base.py#L258). | ||
|
|
||
| The standard input definition looks like this: |
There was a problem hiding this comment.
this is an example, not the definition
zsimjee
reviewed
Aug 14, 2024
|
|
||
| Here, `“text”` contains the text we want to check for toxic language and `“threshold”` contains the confidence threshold for determining toxicity. If the model predicts a toxic level higher than the threshold, then the text is considered toxic. | ||
|
|
||
| The standard output definition looks like this: |
zsimjee
reviewed
Aug 14, 2024
Comment on lines
36
to
37
| # Option 2 - Install only required dependencies (Validator Specific - From README) | ||
| pip install detoxify torch |
There was a problem hiding this comment.
remove this option, mention it as part of #4 here https://github.com/guardrails-ai/guardrails/pull/1008/files#r1717344125
zsimjee
approved these changes
Aug 15, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds a doc explaining how to host validator models.