fluency-eval Workspace – Weights & Biases

Skip to main content

ggMx3KxsNNFGHWp2kjF4X0YeKNZkTbMxHRtLVTzeCPQ

Assets

Assets

All assets Prompts Ops Models Datasets Scorers

accuracy:v0

Name

accuracy(1 version)

Version

0

Last updated

9 months ago

Last updated by

Calls:

1

2

3

4

5

6

7

import weave
@weave.op()
def accuracy(output: dict, flagged: bool):
    flagged_output = output.get("flagged", False)
    return flagged_output == flagged