@@ -3582,6 +3582,7 @@ async def start_trained_model_deployment(
3582
3582
* ,
3583
3583
model_id : str ,
3584
3584
cache_size : t .Optional [t .Union [int , str ]] = None ,
3585
+ deployment_id : t .Optional [str ] = None ,
3585
3586
error_trace : t .Optional [bool ] = None ,
3586
3587
filter_path : t .Optional [t .Union [str , t .Sequence [str ]]] = None ,
3587
3588
human : t .Optional [bool ] = None ,
@@ -3605,6 +3606,7 @@ async def start_trained_model_deployment(
3605
3606
:param cache_size: The inference cache size (in memory outside the JVM heap)
3606
3607
per node for the model. The default value is the same size as the `model_size_bytes`.
3607
3608
To disable the cache, `0b` can be provided.
3609
+ :param deployment_id: A unique identifier for the deployment of the model.
3608
3610
:param number_of_allocations: The number of model allocations on each node where
3609
3611
the model is deployed. All allocations on a node share the same copy of the
3610
3612
model in memory but use a separate set of threads to evaluate the model.
@@ -3631,6 +3633,8 @@ async def start_trained_model_deployment(
3631
3633
__query : t .Dict [str , t .Any ] = {}
3632
3634
if cache_size is not None :
3633
3635
__query ["cache_size" ] = cache_size
3636
+ if deployment_id is not None :
3637
+ __query ["deployment_id" ] = deployment_id
3634
3638
if error_trace is not None :
3635
3639
__query ["error_trace" ] = error_trace
3636
3640
if filter_path is not None :
0 commit comments