Download evals/__init__.py from Kyrlos11/nniback: direct link, hf CLI and curl.
- Browser
- Download file 752 Bytes
-
https://huggingface.co/spaces/Kyrlos11/nniback/resolve/main/evals/__init__.py
- Command line
-
hf download hf://spaces/Kyrlos11/nniback/evals/__init__.py
-
curl -L -o __init__.py https://huggingface.co/spaces/Kyrlos11/nniback/resolve/main/evals/__init__.py
752 Bytes
| """ | |
| Braintrust Evaluation Framework for Kortix Agent. | |
| This module provides tools for evaluating agent performance using Braintrust. | |
| It integrates with the existing agent system to: | |
| 1. Run test cases through the agent | |
| 2. Score outputs using various metrics | |
| 3. Track experiments in Braintrust dashboard | |
| Usage: | |
| # Run evals from command line: | |
| braintrust eval evals/agent_eval.py | |
| # Or programmatically: | |
| from evals import run_agent_eval | |
| await run_agent_eval() | |
| """ | |
| from evals.runner import AgentEvalRunner | |
| from evals.scorers import ( | |
| TaskCompletionScorer, | |
| ToolUsageScorer, | |
| ResponseQualityScorer, | |
| ) | |
| __all__ = [ | |
| "AgentEvalRunner", | |
| "TaskCompletionScorer", | |
| "ToolUsageScorer", | |
| "ResponseQualityScorer", | |
| ] | |