Comments (4)
Hi @firgavin, thanks for reaching us, actually in the #3236 we have a friend raising the similar problem with you, but I didn't reproduce the problem.
If you are using dockerfile exported by 'pf flow build', could you please check the connection file from the output directory? My test output looks like:
And I ran it with docker run -p 8080:8080 -e AZURE_OPEN_AI_CONNECTION_API_KEY=.... name
, the flow can return result successfully.
from promptflow.
@brynn-code Thank you for check this issue. I did search through all open issues but somehow missed that particular one. I've noticed that this feature works correctly only with 'flow serve', as you mentioned in #3236. However, I'm curious why it doesn't work similarly for 'flow test' or 'flow run'? In my scenario, I need to use "flow run" with the same connection management in this container.
from promptflow.
At the early stages of design, we introduced this for 'flow serve' is because we'd like user have a more fluent experience when deploying the flow, as it is frustrated if they need to deploy to several region and then create connection for each one, it will be better if they only have to configure environment variables for each deployment.
For the 'flow test' and 'flow run', usually we assuming that user is doing this locally on their machine, they are still authoring the flow, in this case, we encourage user leverage our connection to store api keys, the stored key was encryted, and encryption key was managed by system ( for example, in windows we will store encrytion key to Credential Manager ), and promptflow package itself will not stored the value anywhere.
Please be careful about it, if user remove it manually, then they can't use connection anymore as promptflow can't decrypt existing connections, unless delete the whole sqlite db.
from promptflow.
Hi, we're sending this friendly reminder because we haven't heard back from you in 30 days. We need more information about this issue to help address it. Please be sure to give us your input. If we don't hear back from you within 7 days of this comment, the issue will be automatically closed. Thank you!
from promptflow.
Related Issues (20)
- [Contribution Request] Correcting some small English grammar mistakes HOT 1
- [Feature Request] Visualizing the evaluations should look different from the promptflow traces, should provide some kind of data visualization HOT 5
- [BUG] [VSCode Extension] Not able to import modules outside of the flow directory HOT 7
- [BUG] KeyError thrown due to environment variables not persisting when calling the promptflow endpoint
- Support remote tracing for CMK-enabled ML workspaces HOT 1
- [BUG]calling pf_client.run inside the callable target function for evaluate got "Error: (AssertionError) daemonic processes are not allowed to have children." HOT 2
- Lisence issue for the dependency on docutils(GPL 3.0) HOT 4
- [BUG] ValueError: Missing required inputs for target : ['question'] while using evalutor_config HOT 3
- [Performance] how to disable all dump, for ex. _node_run_postprocess in run_tracker.py HOT 1
- [BUG] Evaluator respond with Nan values as the first token is a text HOT 1
- [Feature Request] pass addition inputs to target function when using the evaluate method HOT 2
- [Feature Request] Have option to return trace as part of out put of a PromptFlow when deployed as endpoint HOT 12
- [BUG] Running evaluate from promptflow.evals.evaluate and setting trace.destination="none" or "local" causes evaluations to not be available HOT 1
- [BUG] prompt eval method does not take credential and creates PFClient without passing in credential HOT 1
- [BUG] Tracing Contextvar reset HOT 6
- Failed to load trace[BUG] HOT 3
- [BUG] Flow is not enabled HOT 1
- [BUG]evaluator keeps failing with promptflow-eval 0.3.1 but works with 0.3.0 HOT 3
- [BUG] Race condition with global state in process pool HOT 2
- [FeatureAsk] Instantiating a model config for Azure OAI using AAD instead of `api_key` HOT 11
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.