search

LLaVA-VL / LLaVA-Interactive-Demo

LLaVA-Interactive-Demo
https://llava-vl.github.io/llava-interactive/
Apache License 2.0
332 stars 25 forks source link

| ERROR | stderr | KeyError: 'CONTENT_SAFETY_ENDPOINT' #16

Open dengweihuan opened 4 months ago

dengweihuan commented 4 months ago

how to solve the problem of "| ERROR | stderr | KeyError: 'CONTENT_SAFETY_ENDPOINT'" and "KeyError: '"CONTENT_SAFETY_KEY"'". Please tell me what should I set these two variables to.

mattmazzola commented 4 months ago

CONTENT_SAFETY_ENDPOINT and CONTENT_SAFETY_KEY are environment variables you set if you are using any of the Azure AI Content Safety moderation flags (Suffixed with *_aics)

I removed these moderation flags from main branch in this PR in hopes that these variables are only needed if you choose AICS moderation https://github.com/LLaVA-VL/LLaVA-Interactive-Demo/pull/14/files#diff-9561ba98374bfa0ae04e9489e6fcf01cb78e9216b1f31d6322250769bf2e96afL36-L46

To provision Azure AI Content Safety Resource and get the endpoint and key, see the documentation: https://learn.microsoft.com/en-us/azure/ai-services/content-safety/overview