NeMo-Guardrails

NeMo-Guardrails copied to clipboard

Bot intent and Bot Message in response

2

nemoguardrails version 0.7.1 llm-model used = gpt-3.5-turbo getting response having Bot intent: Bot message: sometimes. {'role': 'assistant', 'content': 'Could you please provide more specific details to assist you effectively? This...

praveen062

self_check_facts is not executed

1

im trying to use self_check_facts out of the box here is the relevant part of my yml: rails: output: flows: - self check facts here is colang file: define bot...

AnnPe

Starting actions assigned to variables

1

Hi, thanks for the amazing framework! I am trying to execute an action from a variable, something like this, but i don't know the syntax or if that is possible....

Isaac-AC

TypeError in history_cache_key w hen running basic Langchain chain with guardrails

1

Getting a type error when running the basic langchain with guardrails example. ``` from langchain_core.output_parsers import StrOutputParser template = """ You are a smart assistant that answers questions. Question: {question}...

jordanrfrazier

Can Colang support language other than English?

1

Is it possible to replace "greeting" and the response from English to other language such as Chinese or Japanese? For example. define user express greeting "hello" "hi" define user express...

Lumpylam

Random conversation gets appended before actual input

1

` query="What is the capital of france?" response_rail = await self.rails.generate_async(messages=[{"role": "user", "content": query}])` ``` nemoguardrails.logging.callbacks INFO 04/11/2024 11:37:17.117 - Prompt :: """ Below is a conversation between a helpful...

koszilard

use the consistent delimiter for existing checking and splitting

2

This PR is to fix a corner case. Although the triggering of bug could be rare, the bug in the code needs to be fixed. - in `if "::" in...

HanchenXiong

This PR is in response to issue #451

rgstephens

llm chain not working AttributeError: 'NoneType' object has no attribute 'agenerate_prompt'

2

I have Implemented Neoguardrails in Retrieval QA chain with llama2 model but it is giving error My code: ``` MODEL_TYPE=GPT4All MODEL_PATH=r'C:\Users\komal\Desktop\mages\chatbot\llama-2-7b-chat.Q3_K_M.gguf' MODEL_N_CTX=1000 MODEL_N_BATCH=8 TARGET_SOURCE_CHUNKS=4 llm = LlamaCpp(model_path=MODEL_PATH, n_ctx=MODEL_N_CTX, n_batch=MODEL_N_BATCH, verbose=False)...

Komal-99

pip package installs unexpected module "examples"

1

Hi; I'm unsure whether this is expected but I see that installing the nemo guardrails package installs two additional modules beyond "nemoguardrails": "examples" and "chat-ui". If possible, could those be...

piotrm0