Comments (4)
Hello @chjinche ,
thank you so much for getting back to me and the explanations and sorry for replying only now.
Cedric
from promptflow.
Hi @cedricvidal, both endpoint_name
and deployment_name
support manual input, so you can fill yourself and then can run. Here is my try:
The error you're seeing is just auto listing failure, this is because the tool 'Open Model LLM' depends tightly on azure resources, which means you need to deploy the endpoint
/deployments
in azure portal in advance. To use them in local, you may need to configure azure config first to enable call azure api to get back the endpoints list. Here are detailed steps:
1.Complete the prerequisites mentioned in the tool doc and make sure you deployed the model.
2.Install azure dependencies.
pip install azure-ai-ml
3.Login to azure and set ws triple as default
az login
az account set --subscription <subscription_id>
az configure --defaults group=<resource_group_name> workspace=<workspace_name>
After above setting, you may see the error disappear. For more details about how to connect azure from local, you could see the doc .
Involved the tool contributor @prakharg-msft @dans-msft, could you please add specific instructions of "how to use this tool in local" to 'Open Model LLM' tool doc?
from promptflow.
Thank you @chjinche for the clarification. Indeed, I had run the first two commands but I wasn't aware about the 3rd. Running the following as suggested solved the problem.
az configure --defaults group=<resource_group_name> workspace=<workspace_name>
In addition to adding to the doc, you might want to make it clearer in the Visual editor what the problem is and how to solve it. Or to automatically ask the user to select a default group and workspace if those are not already selected. It would make for an easier experience.
from promptflow.
Hi @cedricvidal it's great to hear that the issue has been resolved! We've implemented a fix to eliminate the misleading error message.
I'd like to emphasize that our Visual Studio Code extension is crafted with the community in mind, ensuring ease of use without the necessity for Azure dependencies. Consequently, we might not be pursuing the addition of new UI features, such as visually selecting Azure workspace information.
'Open Model LLM' tool is special and tightly coupled with azure resource. Our initial approach will be to enhance the tool's documentation with instructions for local usage. Tool owner @dans-msft @prakharg-msft please kindly do take a look and improve the tool doc.
from promptflow.
Related Issues (20)
- [Feature Request] Is there a way to export the trace reports? HOT 1
- Is there a way to set the trace destination in the Python code, rather than via the command line? HOT 13
- Invalid instrumentation key HOT 1
- ImportError: DLL load failed while importing _rust: The specified procedure could not be found HOT 4
- [BUG] Non-retryable server side error: Operation returned an invalid status 'Bad Request'. HOT 1
- [BUG] Evaluate on test dataset using evaluate() with SimilarityEvaluator returns NaN HOT 3
- [BUG] Function `evaluate(...)` from promptflow.evals.evaluate can't be used / imported. HOT 3
- [BUG] pf.config.set failed with MissingAzurePackage: "promptflow[azure]" is required for this functionality, please install it by running "pip install promptflow-azure" with your version. HOT 2
- [BUG] I only want to send 1 trace, but PromptFlow sends 4 every time HOT 6
- [BUG] How do I change the icons in the trace report? HOT 1
- [BUG] Parsing of tool calls does not support arrays in the "arguments"
- [BUG] Cannot connect to promptflow tracing during release deploy pipeline HOT 10
- Bulk processing of inputs in an LLM - bulk categorization HOT 5
- All Blank Input Fields in PromptFlow UI HOT 3
- [FeatureAsk] Unable to use tracing in a promptflow endpoint that has been deployed to Azure Machine Learning HOT 3
- [BUG] Error when running AOAI function calling example HOT 3
- [BUG] Potential incompatible with mlflow>=2.13.1 HOT 1
- Integrate a APIM in front of deployments in PromptFlow HOT 1
- [Feature Request] Hiding Sensitive Inputs HOT 1
- [BUG] Promptflow tracing is not compatible with Azure Web Apps HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from promptflow.