🪢 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 React from "react";
import {
Select,
SelectContent,
SelectItem,
SelectTrigger,
SelectValue,
} from "@/src/components/ui/select";
import { type ScoreChartDataset } from "@/src/features/scores-chart-view/types";
const TRIGGER_CLASS = "h-7 w-auto gap-1 text-xs";
// Same wording the rest of the app already uses for these three views
// (`viewLabels` in `@/src/features/monitors/helpers/monitorLabels`,
// the scores filter groups in `@/src/features/experiments/config/
// experiment-items-filter-config`) — not fresh labels invented here.
const DATASET_OPTIONS: { key: ScoreChartDataset; label: string }[] = [
{ key: "numeric", label: "Scores (numeric)" },
{ key: "boolean", label: "Scores (boolean)" },
{ key: "categorical", label: "Scores (categorical)" },
];
/**
* The "View" picker in the scores chart config panel — which underlying
* scores view (`scores-numeric`, `scores-boolean`, or `scores-categorical`)
* the chart queries.
* Switching this changes which metrics/breakdowns are offered below it (see
* `coerceScoreChartConfig`), since the two views expose different measures.
*/
export const DatasetSelect = React.memo(function DatasetSelect({
value,
onChange,
}: {
value: ScoreChartDataset;
onChange: (value: ScoreChartDataset) => void;
}) {
return (
<Select
value={value}
onValueChange={(v) => onChange(v as ScoreChartDataset)}
>
<SelectTrigger className={TRIGGER_CLASS} aria-label="View">
<SelectValue />
</SelectTrigger>
<SelectContent>
{DATASET_OPTIONS.map((d) => (
<SelectItem key={d.key} value={d.key}>
{d.label}
</SelectItem>
))}
</SelectContent>
</Select>
);
});