Update Classifier
PUThttps://api.confident-ai.com/v1/classifiers/{classifierId}
Updates a classifier. Only the fields you send are changed; omitting a field leaves it untouched, and sending null clears it.
A classifier's dataModel cannot be changed after creation, and a preset can only be applied when creating one. Requires the Starter plan or above.
curl -X PUT "https://api.confident-ai.com/v1/classifiers/{classifierId}" \
-H "CONFIDENT_API_KEY: <PROJECT-API-KEY>" \
-H "Content-Type: application/json" \
-d '{
"autoClassify": false
}'{
"success": true,
"data": {
"classifier": {
"id": "CLASSIFIER-ID",
"name": "Sentiment",
"enabled": true,
"autoClassify": false,
"dataModel": "TRACE",
"labels": []
}
},
"link": "https://app.confident-ai.com/project/<PROJECT-ID>/workflows"
}Headers
CONFIDENT_API_KEYstringRequiredThe API key of your Confident AI project.
Path parameters
classifierIdstringRequiredThe unique identifier of the classifier.
Request body
descriptionstringWhat this classifier is for.
enabledbooleanWhether the classifier runs at all. Defaults to true.
autoClassifybooleanWhether incoming items are classified automatically as they arrive. Defaults to true.
filtersobjectA set of filter groups combined by a top-level operator.
Show 2 propertiesHide 2 properties
operatorenumRequiredHow filters or groups are combined.
Show 2 enum valuesHide 2 enum values
ANDOR
groupslist of objectsRequiredThe filter groups.
Show 2 propertiesHide 2 properties
operatorenumRequiredHow filters or groups are combined.
Show 2 enum valuesHide 2 enum values
ANDOR
filterslist of objectsRequiredThe filter rows in this group.
Show 4 propertiesHide 4 properties
categorystringRequiredThe property a filter row matches on (e.g. "Name", "User Id", "Model", "Metadata"). The set of valid values depends on the line's dataModel.
conditionenumRequiredThe comparison a filter row applies. Valid conditions depend on the category.
Show 18 enum valuesHide 18 enum values
IsIs notIs equal toDoes not equalIs less thanIs equal or less thanIs greater thanIs equal or greater thanHasHas notContainsContains onlyDoes not containHas increased by more thanHas increased by less thanHas decreased by more thanHas decreased by less thanHas changed from
valuestring | number | list of stringsRequiredThe value to match against.
Show 3 variantsHide 3 variants
string
- OR
number
- OR
list of strings
keystringThe property key. Auto-populated from category when omitted; required for Metadata, Metric, and Classifier filters.
autoGenerationConfigobjectDrives label generation. Both required keys must be set before the generate endpoint will run.
Show 3 propertiesHide 3 properties
summaryPromptstringRequiredWhat the model should look for when clustering sampled traffic into themes.
nClustersintegerRequiredRoughly how many themes to cluster the sample into.
sampleSizeintegerHow many traces or threads to sample. Defaults to 200.
namestringA new name, unique per data model within the project.
Response
successbooleanIndicates if the request was successful.
dataobjectShow 1 propertyHide 1 property
classifierobjectShow 9 propertiesHide 9 properties
idstringThe unique identifier of the classifier.
namestringThe classifier's name.
dataModelenumWhat kind of item the classifier classifies.
Show 2 enum valuesHide 2 enum values
TRACETHREAD
labelslist of objectsThe labels this classifier can apply.
Show 6 propertiesHide 6 properties
idstringThe unique identifier of the label.
namestringThe label's name.
descriptionstringWhen this label applies. This is the instruction the classifying model reads, so it should describe the condition rather than restate the name.
enabledbooleanWhether the label can be applied.
statusenumACTIVElabels are in use;RECOMMENDEDones are generated suggestions awaiting review.Show 2 enum valuesHide 2 enum values
RECOMMENDEDACTIVE
polarityenumWhether more of this label is good, bad, or neither, for trend reporting.
Show 3 enum valuesHide 3 enum values
HIGHER_IS_BETTERLOWER_IS_BETTERNEUTRAL
descriptionstringWhat this classifier is for.
enabledbooleanWhether the classifier runs at all. Defaults to true.
autoClassifybooleanWhether incoming items are classified automatically as they arrive. Defaults to true.
filtersobjectA set of filter groups combined by a top-level operator.
Show 2 propertiesHide 2 properties
operatorenumHow filters or groups are combined.
Show 2 enum valuesHide 2 enum values
ANDOR
groupslist of objectsThe filter groups.
Show 2 propertiesHide 2 properties
operatorenumHow filters or groups are combined.
Show 2 enum valuesHide 2 enum values
ANDOR
filterslist of objectsThe filter rows in this group.
Show 4 propertiesHide 4 properties
categorystringThe property a filter row matches on (e.g. "Name", "User Id", "Model", "Metadata"). The set of valid values depends on the line's dataModel.
conditionenumThe comparison a filter row applies. Valid conditions depend on the category.
Show 18 enum valuesHide 18 enum values
IsIs notIs equal toDoes not equalIs less thanIs equal or less thanIs greater thanIs equal or greater thanHasHas notContainsContains onlyDoes not containHas increased by more thanHas increased by less thanHas decreased by more thanHas decreased by less thanHas changed from
valuestring | number | list of stringsThe value to match against.
keystringThe property key. Auto-populated from category when omitted; required for Metadata, Metric, and Classifier filters.
autoGenerationConfigobjectDrives label generation. Both required keys must be set before the generate endpoint will run.
Show 3 propertiesHide 3 properties
summaryPromptstringWhat the model should look for when clustering sampled traffic into themes.
nClustersintegerRoughly how many themes to cluster the sample into.
sampleSizeintegerHow many traces or threads to sample. Defaults to 200.
linkstringA link to the workflows page.