Pamela Fox

You could try out the new integrated vectorization feature! That's all cloud-based indexing, using indexers and skills for chunking and vectorizing. Here's a PR: https://github.com/Azure-Samples/azure-search-openai-demo/pull/1159 You could check out that...

It looks like you did manage to get the branch checked out, but then you ran npm install from the root folder. There's no package.json there, so it failed. The...

@shivam10u I am going to test out the branch myself soon, so I'll see if I have the same error. cc @srbalakr

Are you able to create that deployment from the Azure Portal as well? Sometimes there are availability issues that we haven't accounted for in our Bicep.

Take a look at azure.yaml: https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/azure.yaml It uses "hooks" to execute scripts before or after particular stages. We currently run prepdocs.sh as a postprovision hook, which runs after "azd provision"...

Can you share examples of where it's answering questions that aren't in the provided sources? Have you tried using GPT-4? (It is generally better at sticking to instructions).

Are you all still seeing this? We may move to a slightly tweaked prompt that emphasizes coherence to sources more, you could try this prompt: https://github.com/Azure-Samples/azure-search-openai-demo/pull/967/files#diff-d6c03003d61bd0a593ea552fb76573df32d24fe6ac495da5ac2f7250cc372b96

This is basically the same issue as https://github.com/Azure-Samples/azure-search-openai-demo/issues/431 Please see that issue for some suggestions

We don't have an elegant solution yet, but I think a workaround is to figure out when your token is expiring, and put code in to regenerate the token, such...

A fix has been merged for this error. The fix refreshes the token every 5 minute. Please re-open if you still encounter it during prepdocs.py. Note that you will encounter...