包含: - hair_service_sd: 换发型/换发色算法服务 (端口 8801) - photo_service: LoRA 训练调度服务 (端口 32678) - stable-diffusion-webui: SD WebUI 推理服务 (端口 57860) - kohya_ss_home: 训练环境代码 - meidaojia: 监控测试脚本 - setup.sh: 一键部署脚本 (conda环境恢复 + 配置生成 + 完整性检查) - start_all_services.sh: 启动3个服务 - configure.ini.template: 路径模板化 (BASE_DIR自动推导) - conda_envs/py310.yml: py310 环境定义 大文件 (weights/, models/, data/, conda_envs/*.tar.gz 等) 通过 .gitignore 排除, 由网盘单独上传。
28 lines
694 B
Python
28 lines
694 B
Python
import pynvml
|
|
threshold = 0.9
|
|
def get_gpu(need_gpu_id):
|
|
used = get_gpu_threshold(need_gpu_id)
|
|
#小于一定的
|
|
if used > threshold:
|
|
need_use = get_use_gpu()
|
|
return need_use[0]
|
|
else:
|
|
return need_gpu_id
|
|
|
|
def get_use_gpu():
|
|
use=[]
|
|
for index in range(pynvml.nvmlDeviceGetCount()):
|
|
used = get_gpu_threshold(index)
|
|
if used > threshold:
|
|
use.append(index)
|
|
return use
|
|
|
|
def get_gpu_count():
|
|
return pynvml.nvmlDeviceGetCount()
|
|
|
|
def get_gpu_threshold(index):
|
|
handle = pynvml.nvmlDeviceGetHandleByIndex(index)
|
|
meminfo = pynvml.nvmlDeviceGetMemoryInfo(handle)
|
|
used = meminfo.used / meminfo.total
|
|
return used
|