insights
Get Model Metrics
insights
Get Model Metrics
Calculates f1, precision, and recall for a run/split given a set of filters.
:param macro: Whether to macro average or weighted average metrics. Default False (weighted) :param metrics_request: Filters to pass in before calculating metrics
POST
/
projects
/
{project_id}
/
runs
/
{run_id}
/
split
/
{split}
/
insights
/
metrics
Authorizations
Galileo-API-Key
string
headerrequiredPath Parameters
project_id
string
requiredrun_id
string
requiredsplit
enum<string>
requiredAvailable options:
training, validation, test, inference Query Parameters
inference_name
string
default: macro
boolean
default: falsescope
string | null
default: Body
application/json
task
string | null
filter_params
object
compare_to
enum<string> | null
Available options:
training, validation, test, inference map_threshold
number
default: 0.5meta_cols
string[] | null
Response
200 - application/json
task
string | null
filter_params
object
compare_to
enum<string> | null
Available options:
training, validation, test, inference map_threshold
number
default: 0.5meta_cols
string[] | null
f1
number | null
recall
number | null
precision
number | null
accuracy
number | null
data_error_potential
number | null
confidence
number | null
multi_label_task_metrics
object
Metrics per task for multi-label models.
Was this page helpful?