mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2026-04-23 00:17:25 +08:00
@@ -170,12 +170,7 @@ class PaddleDisWorkerProc:
|
||||
self.worker = get_worker(fd_config=fd_config, local_rank=self.local_rank, rank=self.ranks)
|
||||
|
||||
self.max_chips_per_node = 16 if current_platform.is_iluvatar() else 8
|
||||
self.speculative_decoding = fd_config.speculative_config.method is not None
|
||||
self.enable_overlap_schedule = (
|
||||
current_platform.is_cuda()
|
||||
and self.scheduler_config.enable_overlap_schedule
|
||||
and (not self.speculative_decoding)
|
||||
)
|
||||
self.enable_overlap_schedule = self.scheduler_config.enable_overlap_schedule
|
||||
|
||||
def init_control(self):
|
||||
engine_worker_queue_port = self.parallel_config.local_engine_worker_queue_port
|
||||
|
||||
Reference in New Issue
Block a user