🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
import { type ScoreData } from "./types";
import {
type ScoreDataTypeType,
type ScoreTargetTrace,
type ScoreTarget,
type ScoreTargetSession,
ScoreDataTypeEnum,
} from "@langfuse/shared";
export const isNumericDataType = (dataType: ScoreDataTypeType) =>
dataType === ScoreDataTypeEnum.NUMERIC;
export const isCategoricalDataType = (dataType: ScoreDataTypeType) =>
dataType === ScoreDataTypeEnum.CATEGORICAL;
export const isBooleanDataType = (dataType: ScoreDataTypeType) =>
dataType === ScoreDataTypeEnum.BOOLEAN;
export const isTextDataType = (dataType: ScoreDataTypeType) =>
dataType === ScoreDataTypeEnum.TEXT;
export const isScoreUnsaved = (scoreId?: string | null): boolean => !scoreId;
export const toOrderedScoresList = (list: ScoreData[]): ScoreData[] =>
list.sort((a, b) => a.key.localeCompare(b.key));
export const isTraceScore = (
scoreTarget: ScoreTarget,
): scoreTarget is ScoreTargetTrace => scoreTarget.type === "trace";
export const isSessionScore = (
scoreTarget: ScoreTarget,
): scoreTarget is ScoreTargetSession => scoreTarget.type === "session";
export const formatAnnotateDescription = <Target extends ScoreTarget>(
scoreTarget: Target,
): string => {
let sourceEntity = "session";
if (isTraceScore(scoreTarget)) {
sourceEntity = scoreTarget.observationId ? "observation" : "trace";
}
return `Annotate ${sourceEntity} with scores to capture human evaluation across different dimensions.`;
};
export function formatScoreRange(
min: number | null | undefined,
max: number | null | undefined,
) {
if (min != null && max != null) return `${min} – ${max}`;
if (min != null) return `Min ${min}`;
if (max != null) return `Max ${max}`;
return "Enter a number";
}