> ## Documentation Index > Fetch the complete documentation index at: https://docs.woodwide.ai/llms.txt > Use this file to discover all available pages before exploring further. # Explain Inference Rows > Create an asynchronous explanation job for selected row ids from a completed inference job. Poll the returned explanation job; inline explanations are included on the job detail once it succeeds. ## OpenAPI ````yaml /openapi.json post /jobs/{job_id}/explain openapi: 3.1.0 info: title: Woodwide API description: API service for the Woodwide platform version: 0.1.0 servers: - url: https://api.woodwide.ai description: Production security: [] paths: /jobs/{job_id}/explain: post: tags: - Jobs summary: Explain Inference Rows description: >- Create an asynchronous explanation job for selected row ids from a completed inference job. Poll the returned explanation job; inline explanations are included on the job detail once it succeeds. operationId: explain_job_rows_jobs__job_id__explain_post parameters: - name: job_id in: path required: true schema: type: string title: Job Id requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/JobExplainRequest' responses: '202': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/JobDetail' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: JobExplainRequest: properties: ids: items: anyOf: - type: integer - type: string type: array maxItems: 500 minItems: 1 title: Ids description: Inference row ids to explain. type: object required: - ids title: JobExplainRequest description: Request a per-row explanation job for a completed inference job. JobDetail: properties: id: type: string title: Id description: Public ID of the job (e.g. job_A8K2P9QX). type: type: string title: Type description: 'Job type: dataset_ingest, train, infer_sync, or infer_batch.' status: $ref: '#/components/schemas/JobStatus' description: >- Job status: initializing, pending, queued, running, succeeded, failed, canceled, or rejected. For dataset_ingest jobs, may also be 'waiting_for_upload'. created_at: type: string format: date-time title: Created At started_at: anyOf: - type: string format: date-time - type: 'null' title: Started At finished_at: anyOf: - type: string format: date-time - type: 'null' title: Finished At created_by: anyOf: - additionalProperties: true type: object - type: 'null' title: Created By description: 'Who created this job: {principal_type, user_id, api_key_id}.' actor_name: anyOf: - type: string - type: 'null' title: Actor Name description: >- Display name of who created the job: the user's display name (email fallback), or the API key's name for API-key principals. Populated on dashboard-activity rows; null when the creator is unknown. model_type: anyOf: - type: string - type: 'null' title: Model Type description: >- Model task type when applicable (e.g. prediction, clustering, anomaly). Derived from job input or the linked model; null for jobs without a model context. model_id: anyOf: - type: string - type: 'null' title: Model Id description: >- Public ID of the parent model when the job is tied to a model (training, inference); null otherwise. dataset_id: anyOf: - type: string - type: 'null' title: Dataset Id description: >- Public ID of the dataset when the job references a dataset version (e.g. batch inference on curated data); null for file-only inference and other jobs. dataset_name: anyOf: - type: string - type: 'null' title: Dataset Name description: >- Display name of the linked dataset when resolved via dataset_version; null when there is no dataset or the name is unavailable. input_filename: anyOf: - type: string - type: 'null' title: Input Filename description: >- Original name of the uploaded file for file-backed jobs (sync/async file inference); null when the job referenced a dataset or carries no uploaded file. parent_job_id: anyOf: - type: string - type: 'null' title: Parent Job Id description: >- Public ID of the parent job when this job consumes outputs from it as part of a compound inference/training chain. input: anyOf: - additionalProperties: true type: object - type: 'null' title: Input output: anyOf: - additionalProperties: true type: object - type: 'null' title: Output resources: anyOf: - $ref: '#/components/schemas/IngestResources' - $ref: '#/components/schemas/TrainResources' - $ref: '#/components/schemas/InferSyncResources' - $ref: '#/components/schemas/InferBatchResources' - $ref: '#/components/schemas/ModelExplainResources' - type: 'null' title: Resources description: Typed resource references for this job. explanations: anyOf: - items: $ref: '#/components/schemas/JobExplanationRow' type: array - type: 'null' title: Explanations description: Inline row explanations for succeeded infer_explain jobs. error_message: anyOf: - type: string - type: 'null' title: Error Message description: Human-readable error message if the job failed. woodwide_runtime_version: anyOf: - type: string - type: 'null' title: Woodwide Runtime Version description: ML runtime version used for this job. credits_consumed: anyOf: - type: number - type: 'null' title: Credits Consumed description: Credits consumed by this job (null if not yet finalized). progress: anyOf: - type: number - type: 'null' title: Progress description: >- Progress percentage (0-100). For training jobs, tracks training progress. current_step: anyOf: - type: integer - type: 'null' title: Current Step description: >- Current training step (training jobs only). Counts SSL + finetune steps. total_steps: anyOf: - type: integer - type: 'null' title: Total Steps description: Total training steps planned for this job (training jobs only). current_phase: anyOf: - type: string enum: - ssl - finetune - type: 'null' title: Current Phase description: >- Current training phase: ``ssl`` (self-supervised pre-training) or ``finetune`` (supervised head finetune). Training jobs only; null for unsupervised model types that have no training loop. metrics_history: anyOf: - items: $ref: '#/components/schemas/MetricPoint' type: array - type: 'null' title: Metrics History description: >- Bounded series of per-interval training metrics (capped at 200 points, downsampled in place). Training jobs only; null for unsupervised paths and for jobs before the first validation gate. type: object required: - id - type - status - created_at title: JobDetail description: |- Detailed job information. Extends ``Job`` with input/output payloads, typed resource references, error details, and billing information. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError JobStatus: type: string enum: - initializing - pending - queued - running - succeeded - failed - canceled - rejected - waiting_for_upload title: JobStatus IngestResources: properties: type: type: string const: dataset_ingest title: Type default: dataset_ingest dataset: $ref: '#/components/schemas/DatasetVersionRef' type: object required: - dataset title: IngestResources description: Resources for a dataset_ingest job. TrainResources: properties: type: type: string const: train title: Type default: train model: $ref: '#/components/schemas/ModelVersionRef' dataset: $ref: '#/components/schemas/DatasetVersionRef' type: object required: - model - dataset title: TrainResources description: Resources for a train job. InferSyncResources: properties: type: type: string const: infer_sync title: Type default: infer_sync model: $ref: '#/components/schemas/ModelVersionRef' type: object required: - model title: InferSyncResources description: Resources for a synchronous inference job. InferBatchResources: properties: type: type: string const: infer_batch title: Type default: infer_batch model: $ref: '#/components/schemas/ModelVersionRef' dataset: anyOf: - $ref: '#/components/schemas/DatasetVersionRef' - type: 'null' type: object required: - model title: InferBatchResources description: |- Resources for a batch inference job. ``dataset`` is null when the job ran on an uploaded file rather than a curated dataset version (async file inference). ModelExplainResources: properties: type: type: string const: model_explain title: Type default: model_explain model: $ref: '#/components/schemas/ModelVersionRef' dataset: $ref: '#/components/schemas/DatasetVersionRef' type: object required: - model - dataset title: ModelExplainResources description: Resources for a model-scoped explanation job. JobExplanationRow: properties: id: type: string title: Id explanation_text: anyOf: - type: string - type: 'null' title: Explanation Text explanation_label: anyOf: - type: string - type: 'null' title: Explanation Label explanation_gloss: anyOf: - type: string - type: 'null' title: Explanation Gloss feature_contributions: anyOf: - items: additionalProperties: true type: object type: array - type: 'null' title: Feature Contributions prediction_flip: anyOf: - additionalProperties: true type: object - type: 'null' title: Prediction Flip type: object required: - id title: JobExplanationRow description: One row explanation returned by an explanation job. MetricPoint: properties: step: type: integer title: Step description: Training step at which this point was sampled. train_loss: anyOf: - type: number - type: 'null' title: Train Loss description: Average training loss over the interval ending at ``step``. val_loss: anyOf: - type: number - type: 'null' title: Val Loss description: Validation loss at ``step``. Null when no val gate fired. type: object required: - step title: MetricPoint description: |- One per-interval training metrics sample. Emitted by the trainer at each validation gate; the worker handler appends these to ``job.output_json['metrics_history']`` (capped at 200 points with in-place every-other downsampling). ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError DatasetVersionRef: properties: id: type: string title: Id description: Public ID of the parent entity (model or dataset). version_id: type: string title: Version Id description: Public ID of the referenced version. version_number: type: integer title: Version Number archived: type: boolean title: Archived description: True if this dataset version has been archived (soft-deleted). default: false type: object required: - id - version_id - version_number title: DatasetVersionRef description: Dataset version reference. ModelVersionRef: properties: id: type: string title: Id description: Public ID of the parent entity (model or dataset). version_id: type: string title: Version Id description: Public ID of the referenced version. version_number: type: integer title: Version Number type: object required: - id - version_id - version_number title: ModelVersionRef description: Model version reference. securitySchemes: HTTPBearer: type: http scheme: bearer ````