apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

fix typos (#3684)

Browse Source
This commit is contained in:
co63oc
2025-09-01 17:50:17 +08:00
committed by GitHub
parent 0513a78ecc
commit d6369b4d51
67 changed files with 85 additions and 85 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
fastdeploy/worker/gpu_worker.py
View File

@@ -60,7 +60,7 @@ class GpuWorker(WorkerBase):
"""
self.max_chips_per_node = 16 if current_platform.is_iluvatar() else 8
if self.device_config.device_type == "cuda" and paddle.device.is_compiled_with_cuda():
# Set evironment variable
# Set environment variable
self.device_ids = self.parallel_config.device_ids.split(",")
self.device = f"gpu:{self.local_rank % self.max_chips_per_node}"
paddle.device.set_device(self.device)
@@ -169,7 +169,7 @@ class GpuWorker(WorkerBase):
)
)
return available_kv_cache_memory # return to caculate the block num in this device
return available_kv_cache_memory # return to calculate the block num in this device
def load_model(self) -> None:
"""Load model"""
@@ -209,7 +209,7 @@ class GpuWorker(WorkerBase):
"""
if self.model_runner.graph_opt_level >= 1:
self.model_runner.sot_warmup()
# Triger cuda grpah capture
# Trigger cuda graph capture
self.model_runner.capture_model()
def check_health(self) -> bool: