Create Pseudo Label Job
POST/api/2/auto_eval/pseudo_label_job/create
Step 1 of 2: Create a new job configuration for LLM Judge labeling, to be subsequently submitted by calling SubmitPseudoLabelJob.
Request
- application/json
Body
Array [
]
pseudoLabelJobConfig
object
required
Partial configuration containing updates via its non-null fields.
Optional name for the job.
Optional description for the job.
ID of the main dataset to be pseudo-labeled
ID of the dataset containing few-shot examples. Optional.
ID of the actively labeled dataset. Optional. If null, this job is for active learning.
chatCompletionConfig
object
required
For Chat LLM based labeling, the configuration to use with the requests (messages omitted)
The ID of the model to use for the completion.
messages
object[]
required
Role can be 'system', 'user', or 'assistant'.
The content of the message.
The maximum number of tokens to generate.
The temperature to use for the completion.
The top_p value to use for the completion.
promptTemplate
object
required
The template string that defines the prompt
Reserved field. Do not use at the moment.
If true, skip active labeling, which involves an intermediate Dataset created for human labeling.
Responses
- 200
Successful operation
- application/json
- Schema
- Example (from schema)
Schema
{
"jobId": "string"
}