OpenCSGs
Create a new fine-tuning instance, how to use local GPUs resources
suppose you have managed GPUs with k8s, please copy the kube config file to .kube folder of service csghub_server_runner.
.kube folder mounted in docker compose yaml config:
csghub_server_runner:
...
volumes:
- ./.kube:/root/.kube:r
resource list are read from table space_resources, change the config according to your real GPU instances.
资源列表从表中读取
space_resources，根据您的实际 GPU 实例更改配置。
Are there any standards for modification? For example, the GPU instance used in my k8s cluster is NVIDIA-GeForce-RTX-4070 4 cards, which are managed uniformly using gpu-operator. After adding relevant information, it still cannot be selected. The space_resources table configuration is shown in the figure below
资源列表从表中读取，根据您的实际 GPU 实例更改配置。
space_resources是否有任何修改标准？例如，我的 k8s 集群中使用的 GPU 实例是 NVIDIA-GeForce-RTX-4070 4 卡，使用 gpu-operator 统一管理。添加相关信息后，仍然无法选择。space_resources表配置如下图所示
I'm also facing this issue. It seems like modifying SQL alone won't work. Have you solved this problem yet
@jksj-223 @tianj0522 please reduce the number of vcpu, maybe the cpu is not enough to run.
@tianj0522 I’m having the same issue. Did you eventually solve it?
make sure you have followed this example: