TensorFlow

API

 tf.compat / v1 / v1.ConfigProto.DeviceCountEntryConfigProto.DeviceCountEntry


A ProtocolMessage

collective_deterministic_sequential_execution bool collective_deterministic_sequential_execution
collective_group_leader string collective_group_leader
collective_nccl bool collective_nccl
disable_output_partition_graphs bool disable_output_partition_graphs
disable_thread_spinning bool disable_thread_spinning
enable_mlir_bridge bool enable_mlir_bridge
enable_mlir_graph_optimization bool enable_mlir_graph_optimization
executor_type string executor_type
mlir_bridge_rollout MlirBridgeRollout mlir_bridge_rollout
optimize_for_static_graph bool optimize_for_static_graph
recv_buf_max_chunk int32 recv_buf_max_chunk
session_metadata SessionMetadata session_metadata
share_cluster_devices_in_session bool share_cluster_devices_in_session
share_session_state_in_clusterspec_propagation bool share_session_state_in_clusterspec_propagation
use_numa_affinity bool use_numa_affinity
xla_fusion_autotuner_thresh int64 xla_fusion_autotuner_thresh

MLIR_BRIDGE_ROLLOUT_DISABLED 2
MLIR_BRIDGE_ROLLOUT_ENABLED 1
MLIR_BRIDGE_ROLLOUT_UNSPECIFIED 0
MlirBridgeRollout


此页内容是否对您有帮助