toloka.client.pool.dynamic_overlap_config.DynamicOverlapConfig | Source code
DynamicOverlapConfig( self, *, type: Union[Type, str, None] = None, max_overlap: Optional[int] = None, min_confidence: Optional[float] = None, answer_weight_skill_id: Optional[str] = None, fields: Optional[List[Field]] = None)Dynamic overlap settings.
Toloka can automatically increase an overlap of tasks if the confidence level of aggregated responses is not high enough.
Dynamic overlap uses the BASIC algorithm.
Each response is assigned a weight depending on the Toloker's skill value.
The aggregated response confidence is calculated based on the probability algorithm.
The task overlap increases until it reaches max_overlap or until the confidence of the aggregated response exceeds min_confidence.
Note, that if you use dynamic overlap, then set the auto_close_after_complete_delay_seconds pool parameter to a non zero value.
Learn more about the Dynamic overlap in the guide.
| Parameters | Type | Description |
|---|---|---|
type | Optional[Type] | The dynamic overlap algorithm. |
max_overlap | Optional[int] | Maximum overlap. The value must be higher than the default overlap value. Allowed range: from 1 to 30,000. |
min_confidence | Optional[float] | Minimum required confidence of the aggregated response. Allowed range: from 0 to 1. |
answer_weight_skill_id | Optional[str] | A skill that determines the weight of the Toloker's responses. For the best results, use a skill calculated as a percentage of correct responses in control tasks. |
fields | Optional[List[Field]] | A list of output data fields used for aggregating responses. For the best results, each field must have a limited number of response options. Don't specify fields in the list together if they depend on each other. |
Last updated: August 28, 2023