NeMo-Guardrails
NeMo-Guardrails copied to clipboard
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.
This commit introduces a search threshold to the `BasicEmbeddingsIndex` class. This threshold is used to filter the search results based on their distances. The `_filter_results` method is added to perform...
- Added global context variables in guardrails.co. - Implemented flow name validation for Colang 2.x. - Migrated the library to support Colang 2.x.
This PR allows colang2.0-alpha migration. It also refactors the migration.py and adds `from_version` to CLI. Remaining Tasks: - [x] Should we save the old version of the co files? It...
This PR resolves #589
I am trying to establish guardrails setup for Azure OpenAI models. I have used the below configuration in config.yaml. models: -type: main engine: azure model: gpt-4 parameters: azure_endpoint: "" api_version:...
I've been trying to use Nemo-Guardrails with Azure but keep getting the dreaded `"openai.error.InvalidRequestError: Resource not found"` error. Outside of Nemo-Guardrails, my Azure credentials work with `openai python api `as...
llm_params from GenerationOptions are not considered during for the following operations: generate_user_intent, generate_next_step, generate_value
I have an AWS Lambda function using Guardrails and on my handler file I do already run ``` logging.basicConfig(level=logging.DEBUG) ``` However I am not getting log messages beyond WARNING and...
Input: ```json { "config_id": "config", "messages": [ { "role": "user", "content": "How to cook egg\n" } ] } ``` Logs: ``` Event UtteranceUserActionFinished {'final_transcript': 'How to cook egg\n'} Event StartInternalSystemAction...