Question configuration for an Ordinal Score evaluation. **Signature:** export declare interface ScoreQuestion ## Properties | Property | Modifiers | Type | Description | |---|---|---|---| | [rubric](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestionrubric) | | string\[\] | List of ordered rubric grade descriptors from lowest to highest. | | [options](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestionoptions) | | ClassifierOption\[\] | Ordered list of rubric levels with `{ label, description }`. When all `label` values parse as finite numbers (e.g. "1".."5"), `expectedScore` is computed directly over those numeric level values. | | [temperature](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestiontemperature) | | number | Softmax temperature override (\> 0.0). When unset or \<= 0, the model's calibrated temperature is used. | | [instructions](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestioninstructions) | | string | Optional task instruction prompt for the model. | | [context](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestioncontext) | | string | Optional domain or schema context prepended to instructions. | | [imageBytes](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestionimagebytes) | | Uint8Array | Optional raw encoded image bytes (PNG, JPEG, WebP, or TGA) for the context. | | [audioSamples](https://developers.google.com/edge/api/mediapipe/js/tasks-decision.scorequestion#scorequestionaudiosamples) | | Float32Array | Optional mono 16kHz PCM audio samples in \[-1.0, 1.0\] for the context. | ## ScoreQuestion.rubric {#scorequestionrubric} List of ordered rubric grade descriptors from lowest to highest. **Signature:** rubric?: string[]; ## ScoreQuestion.options {#scorequestionoptions} Ordered list of rubric levels with `{ label, description }`. When all `label` values parse as finite numbers (e.g. "1".."5"), `expectedScore` is computed directly over those numeric level values. **Signature:** options?: ClassifierOption[]; ## ScoreQuestion.temperature {#scorequestiontemperature} Softmax temperature override (\> 0.0). When unset or \<= 0, the model's calibrated temperature is used. **Signature:** temperature?: number; ## ScoreQuestion.instructions {#scorequestioninstructions} Optional task instruction prompt for the model. **Signature:** instructions?: string; ## ScoreQuestion.context {#scorequestioncontext} Optional domain or schema context prepended to instructions. **Signature:** context?: string; ## ScoreQuestion.imageBytes {#scorequestionimagebytes} Optional raw encoded image bytes (PNG, JPEG, WebP, or TGA) for the context. **Signature:** imageBytes?: Uint8Array; ## ScoreQuestion.audioSamples {#scorequestionaudiosamples} Optional mono 16kHz PCM audio samples in \[-1.0, 1.0\] for the context. **Signature:** audioSamples?: Float32Array;