Skip to main content
Quality targets define measurable quality goals for your labeling projects. They can trigger alerts when quality metrics fall below specified thresholds.

​
Methods

​
list

List all quality targets for a project. 
client.quality_targets.list(
    project_uid="proj_123",
    limit=100,
    cursor="next_page_token"
)
​
project_uid
str
required
UID of the project
​
limit
int
Maximum number of quality targets to return per page
​
cursor
str
Pagination cursor for fetching the next page
​
data
list[QualityTarget]
List of quality target objects
​
next_cursor
str
Cursor for the next page of results

​
get

Retrieve a specific quality target. 
client.quality_targets.get(
    project_uid="proj_123",
    uid="target_123"
)
​
project_uid
str
required
UID of the project
​
uid
str
required
Unique identifier of the quality target
​
uid
str
Unique identifier of the quality target
​
name
str
Name of the quality target
​
metric
str
Quality metric to measure (e.g., “accuracy”, “precision”, “recall”)
​
threshold
float
Target threshold value
​
operator
str
Comparison operator (e.g., “gte”, “lte”, “eq”)
​
severity
str
Severity level when threshold is not met (e.g., “warning”, “error”)
​
is_active
bool
Whether the quality target is currently active
​
notify_webhook
bool
Whether to send webhook notifications
​
notify_emails
list[str]
Email addresses to notify when threshold is not met

​
create

Create a new quality target for a project. 
client.quality_targets.create(
    project_uid="proj_123",
    name="Minimum Accuracy Target",
    metric="accuracy",
    threshold=0.95,
    operator="gte",
    severity="warning",
    is_active=True,
    notify_webhook=True,
    notify_emails=["[email protected]"]
)
​
project_uid
str
required
UID of the project
​
name
str
required
Name of the quality target
​
metric
str
required
Quality metric to measure (e.g., “accuracy”, “precision”, “recall”, “f1_score”)
​
threshold
float
required
Target threshold value (0.0 to 1.0 for most metrics)
​
operator
str
Comparison operator: “gte”, “lte”, “eq”, “gt”, or “lt”
​
severity
str
Severity level: “info”, “warning”, “error”, “critical”
​
is_active
bool
Whether the quality target should be active immediately (default: true)
​
notify_webhook
bool
Whether to send webhook notifications when threshold is not met
​
notify_emails
list[str]
Email addresses to notify when threshold is not met

​
update

Update an existing quality target. 
client.quality_targets.update(
    project_uid="proj_123",
    uid="target_123",
    threshold=0.98,
    severity="error"
)
​
project_uid
str
required
UID of the project
​
uid
str
required
Unique identifier of the quality target to update
​
name
str
Updated name
​
metric
str
Updated metric
​
threshold
float
Updated threshold value
​
operator
str
Updated comparison operator
​
severity
str
Updated severity level
​
is_active
bool
Whether the quality target should be active
​
notify_webhook
bool
Updated webhook notification setting
​
notify_emails
list[str]
Updated email notification list

​
delete

Delete a quality target. 
client.quality_targets.delete(
    project_uid="proj_123",
    uid="target_123"
)
​
project_uid
str
required
UID of the project
​
uid
str
required
Unique identifier of the quality target to delete

​
evaluate

Evaluate all quality targets for a project against current metrics. 
client.quality_targets.evaluate(project_uid="proj_123")
​
project_uid
str
required
UID of the project to evaluate
​
results
list[QualityTargetEvaluation]
List of evaluation results for each quality target
​
results[].target_uid
str
UID of the evaluated quality target
​
results[].metric
str
Metric that was evaluated
​
results[].current_value
float
Current value of the metric
​
results[].threshold
float
Target threshold
​
results[].passed
bool
Whether the current value meets the threshold
​
results[].severity
str
Severity level if threshold not met

​
Async Methods

All methods are available in async form through AsyncAvala: 
await client.quality_targets.list(project_uid="proj_123")
await client.quality_targets.get(project_uid="proj_123", uid="target_123")
await client.quality_targets.create(
    project_uid="proj_123",
    name="Accuracy Target",
    metric="accuracy",
    threshold=0.95
)
await client.quality_targets.update(
    project_uid="proj_123",
    uid="target_123",
    threshold=0.98
)
await client.quality_targets.delete(project_uid="proj_123", uid="target_123")
await client.quality_targets.evaluate(project_uid="proj_123")

Build docs developers (and LLMs) love

Get started for freeTalk to us