Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[remote] Create RemoteSession class that is available from a notebook setting #2136

Closed
Tracked by #1689
mikeldking opened this issue Jan 26, 2024 · 1 comment · Fixed by #2151
Closed
Tracked by #1689

[remote] Create RemoteSession class that is available from a notebook setting #2136

mikeldking opened this issue Jan 26, 2024 · 1 comment · Fixed by #2151
Assignees
Labels
c/dx Developer experience c/evals c/traces enhancement New feature or request

Comments

@mikeldking
Copy link
Contributor

mikeldking commented Jan 26, 2024

As a user I still want to be able to interact with phoenix in a notebook even if it's running in a remote server. As a user I want to:

  • have access to traces
  • have access to evals
  • be able to send evals

The UX for interacting with a remote session should ideally be no different than a local one

Copy link

dosubot bot commented Jan 26, 2024

🤖

Based on the current implementation of Arize Phoenix, it seems that the functionality you're requesting is partially available. The Session class, which maintains a 1-1 shared state with the Phoenix App, already supports specifying a remote host and port. This means that you can run the Phoenix App on a remote server and interact with it from a notebook.

However, the current implementation assumes that the datasets (including traces) are available locally when the Session is initialized. If your datasets are also stored on the remote server, you would need to load them into your local environment before initializing the Session.

Here's a simplified example of how you might do this:

# Load datasets from remote server
primary_dataset = load_dataset_from_remote('primary_dataset_path')
trace_dataset = load_trace_dataset_from_remote('trace_dataset_path')

# Initialize session with remote host and port
session = Session(
    primary_dataset=primary_dataset,
    trace_dataset=trace_dataset,
    host='remote_host',
    port='remote_port'
)

In this example, load_dataset_from_remote and load_trace_dataset_from_remote are hypothetical functions that you would need to implement. These functions should download the datasets from the remote server and return them in a format that can be passed to the Session constructor.

As for sending evaluations, the current implementation does not seem to support this directly. However, you could potentially implement this functionality by extending the Session class and adding a method for sending evaluations to the remote server.

Please note that these are just suggestions and might require further adjustments based on the specifics of your remote server and how you have set up your Arize Phoenix environment.

For more details, you can refer to the Session class implementation.

I hope this helps! If you have further questions or if something is not clear, please let me know.

Sources


This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

@dosubot dosubot bot added c/dx Developer experience c/evals c/traces enhancement New feature or request labels Jan 26, 2024
@RogerHYang RogerHYang moved this from 📘 Todo to 🔍. Needs Review in phoenix Jan 31, 2024
@github-project-automation github-project-automation bot moved this from 🔍. Needs Review to ✅ Done in phoenix Feb 5, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
c/dx Developer experience c/evals c/traces enhancement New feature or request
Projects
Archived in project
Development

Successfully merging a pull request may close this issue.

2 participants