This is an automated email from the ASF dual-hosted git repository. yongzao pushed a commit to branch deploy in repository https://gitbox.apache.org/repos/asf/iotdb.git
commit e44936f8a0732377305668e9ad7dcfacb7b51644 Author: Yongzao <[email protected]> AuthorDate: Fri Aug 29 13:07:14 2025 +0800 stash --- .../ainode/core/inference/pool_scheduler/basic_pool_scheduler.py | 2 +- iotdb-core/ainode/ainode/core/manager/inference_manager.py | 4 +--- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/iotdb-core/ainode/ainode/core/inference/pool_scheduler/basic_pool_scheduler.py b/iotdb-core/ainode/ainode/core/inference/pool_scheduler/basic_pool_scheduler.py index 8994a9e00b2..0343c2a1a52 100644 --- a/iotdb-core/ainode/ainode/core/inference/pool_scheduler/basic_pool_scheduler.py +++ b/iotdb-core/ainode/ainode/core/inference/pool_scheduler/basic_pool_scheduler.py @@ -51,7 +51,7 @@ class BasicPoolScheduler(AbstractPoolScheduler): Schedule a scaling action for the given model_id. """ if model_id not in self._request_pool_map: - pool_num = estimate_pool_size(self.DEFAULT_DEVICE, model_id) + pool_num = 2 if pool_num <= 0: raise InferenceModelInternalError( f"Not enough memory to run model {model_id}." diff --git a/iotdb-core/ainode/ainode/core/manager/inference_manager.py b/iotdb-core/ainode/ainode/core/manager/inference_manager.py index 2d4e2088ac7..a189274595b 100644 --- a/iotdb-core/ainode/ainode/core/manager/inference_manager.py +++ b/iotdb-core/ainode/ainode/core/manager/inference_manager.py @@ -242,9 +242,7 @@ class InferenceManager: predict_length, ) - if model_id in self.ACCELERATE_MODEL_ID and "cuda" in str( - self.DEFAULT_DEVICE - ): + if model_id in self.ACCELERATE_MODEL_ID: # TODO: Logic in this branch shall handle all LTSM inferences # TODO: TSBlock -> Tensor codes should be unified data = full_data[1][0]
