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

feat(ui): Very basic MVP for viewing a Scorer's logs #3556

Open
wants to merge 22 commits into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
493 changes: 493 additions & 0 deletions docs/docs/guides/evaluation/builtin_scorers.mdx

Large diffs are not rendered by default.

19 changes: 11 additions & 8 deletions docs/docs/guides/evaluation/guardrails_and_monitors.md
Original file line number Diff line number Diff line change
Expand Up @@ -23,11 +23,11 @@ Throughout this guide, we'll refer to functions decorated with `@weave.op` as "o
:::

#### Ready-to-Use Scorers
While this guide shows you how to create custom scorers, Weave comes with a variety of [predefined scorers](./scorers.md#predefined-scorers) that you can use right away, including:
- [Hallucination detection](./scorers.md#hallucinationfreescorer)
- [Summarization quality](./scorers.md#summarizationscorer)
- [Embedding similarity](./scorers.md#embeddingsimilarityscorer)
- [Relevancy evaluation](./scorers.md#ragas---contextrelevancyscorer)
While this guide shows you how to create custom scorers, Weave comes with a variety of [predefined scorers](./builtin_scorers.mdx) that you can use right away, including:
- [Hallucination detection](./builtin_scorers.mdx#hallucinationfreescorer)
- [Summarization quality](./builtin_scorers.mdx#summarizationscorer)
- [Embedding similarity](./builtin_scorers.mdx#embeddingsimilarityscorer)
- [Relevancy evaluation](./builtin_scorers.mdx#ragas---contextrelevancyscorer)
- And more!


Expand Down Expand Up @@ -307,6 +307,12 @@ score = scorer.score(output="some text")
- ✅ No Op required
- ❌ No association with the LLM/Op call

### Score Analysis


For detailed information about querying calls and their scorer results, see our [Score Analysis Guide](./scorers.md#score-analysis) and our [Data Access Guide](/guides/tracking/tracing#querying--exporting-calls).


## Production Best Practices

### 1. Set Appropriate Sampling Rates
Expand Down Expand Up @@ -340,14 +346,11 @@ async def evaluate_comprehensively(call):
- Set up alerts for concerning patterns (coming soon)

### 4. Access Historical Data

Scorer results are stored with their associated calls and can be accessed through:
- The Call object's `feedback` field
- The Weave Dashboard
- Our query APIs

For detailed information about querying calls and their scorer results, see our [Data Access Guide](/guides/tracking/tracing#querying--exporting-calls).

### 5. Initialize Guards Efficiently

For optimal performance, especially with locally-run models, initialize your guards outside of the main function. This pattern is particularly important when:
Expand Down
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading