README.md

metadata

benchmark: ttt
task: xxx
type: prediction

Batch job

model_id: {model_id} dataset_name: {job.dataset_name} dataset_config: {job.dataset_config} dataset_split: {job.dataset_split} dataset_column: {job.dataset_column}