yiluxiangbei
Loading Heatmap…

yiluxiangbei created repository ai/octopus

2 months ago

yiluxiangbei created repository ai/octopus-charts

3 months ago

yiluxiangbei created repository ai/octopus-doc

3 months ago

yiluxiangbei created repository yiluxiangbei/octopus-doc

7 months ago

yiluxiangbei commented on issue OpenI/octopus#154

公共算法是谁来维护上传?

> 目前这种多变的业务场景,我们会归纳于章鱼与第三方平台对接的场景,从技术架构上来说,目前`openai-server`和`admin-server`作为章鱼内部的业务接口网关服务定义了章鱼内部的业务接口,而我们计划开发`third-server`来实现平台之间业务的实现,这一块接口采用专门的授权认证比如`oauth2.0`,定制自由都更高的接口比如支持存储挂载的定义,你认为这种方式会有帮助么 test

10 months ago

yiluxiangbei commented on issue OpenI/octopus#154

公共算法是谁来维护上传?

有几个角色,算法使用者,算法开发者,调度平台(本项目),算法商店;算法商店连接算法使用者和算法开发者,算法开发者可以通过使用者购买获得收益

10 months ago

yiluxiangbei commented on issue OpenI/octopus#154

公共算法是谁来维护上传?

> > > 用户从我的算法里面点击分享到公共算法中 > > > > > > Ok, 不过建议做成类似商店的模式,独立于平台,在第三方平台可以买之类的 > > > > @shamartor 怎么看? > > 我们目前的一些设计主要基于一些科研机构内部的科研团队场景的需求,你说的这个我觉得很有意思,这是你那边的应用场景么?是否能从使用者或者使用场景来描述一下,我们再来进一步分析讨论 test

10 months ago

yiluxiangbei commented on issue OpenI/octopus#157

模型能力开放什么时候支持?

> > 使用容器起http服务,选择模型创建容器服务 > > 请问你这边需要支持模型部署的场景主要是为了验证模型还是为了提供稳定的在线模型推理服务呢?会比较注重哪些机器学习平台框架的模型呢? 应该提供不同的容器,或者自己创建模型开放容器,选择模型,选择容器,部署

10 months ago

yiluxiangbei commented on issue OpenI/octopus#157

模型能力开放什么时候支持?

使用容器起http服务,选择模型创建容器服务

10 months ago

yiluxiangbei opened issue OpenI/octopus#157

模型能力开放什么时候支持?

10 months ago

yiluxiangbei commented on issue OpenI/octopus#154

公共算法是谁来维护上传?

> 用户从我的算法里面点击分享到公共算法中 Ok, 不过建议做成类似商店的模式,独立于平台,在第三方平台可以买之类的

10 months ago

yiluxiangbei created repository deeplearning/octopus-charts

10 months ago

yiluxiangbei created repository deeplearning/octopus-doc

10 months ago

yiluxiangbei opened issue OpenI/octopus#154

公共算法是谁来维护上传?

10 months ago

yiluxiangbei commented on issue OpenI/octopus#152

can't get taskmetrics when using "npm run dev", the url is this.href

> 本地调试执行npm run dev指令需要修改DOMAIN地址,将http://192.168.202.73替换成自己的服务器地址 > 具体步骤为管理端:打开octopus->admin-portal->src->api->globalVariable.js > ![url路径问题](/attachments/54642606-17dc-4d94-a839-6bb0c3d39060) > 管理端:打开octopus->openai-portal->src->api->globalVariable.js > > ![url路径问题](/attachments/d42f4d2c-7fd2-4154-94f2-0e4835c0171b) > 呵呵

10 months ago

yiluxiangbei commented on issue OpenI/octopus#151

get train log with 404, No such file or directory

> > I think the reason is that the path "/data" is mount from another disk. And the filebeat container has no "/data" path. But the docker path /var/lib/docker links to /data/var/lib/docker > > Was it caused by this problem? How did you solve it in the end? We want to understand clearly so that this project can do better,thx! The container has no path /data, but the container use host path /var/log/containers. I add /data to the container, it can work well.

10 months ago

yiluxiangbei commented on issue OpenI/octopus#151

get train log with 404, No such file or directory

It can work now.

10 months ago

yiluxiangbei commented on issue OpenI/octopus#151

get train log with 404, No such file or directory

Where is filebeat.yml?

10 months ago

yiluxiangbei commented on issue OpenI/octopus#151

get train log with 404, No such file or directory

/data/root/nfs_root/data/logger The image is broadinstitute/gatk. ``` docker pull broadinstitute/gatk ``` see https://github.com/broadinstitute/gatk. try this command ``` Run a tool: ./gatk PrintReads -I src/test/resources/NA12878.chr17_69k_70k.dictFix.bam -O output.bam ```

10 months ago

yiluxiangbei commented on issue OpenI/octopus#151

get train log with 404, No such file or directory

> Can refer to [here](https://kubernetes.io/docs/concepts/cluster-administration/logging/#logging-at-the-node-level),the log needs to be output to stdout and stderr in container. kubectl logs has data, but no log file in logger path. kubectl logs -f yc6e593b92ea46fd9ee188afd091cf7d-task0-0 -n d37c68f328b94e8da56bde693916a55e ``` 09:10:15.094 INFO NativeLibraryLoader - Loading libgkl_compression.so from jar:file:/gatk/gatk-package-4.2.2.0-local.jar!/com/intel/gkl/native/libgkl_compression.so Nov 05, 2021 9:10:15 AM shaded.cloud_nio.com.google.auth.oauth2.ComputeEngineCredentials runningOnComputeEngine INFO: Failed to detect whether we are running on Google Compute Engine. 09:10:15.271 INFO PrintReads - ------------------------------------------------------------ 09:10:15.272 INFO PrintReads - The Genome Analysis Toolkit (GATK) v4.2.2.0 09:10:15.272 INFO PrintReads - For support and documentation go to https://software.broadinstitute.org/gatk/ 09:10:15.272 INFO PrintReads - Executing as root@yc6e593b92ea46fd9ee188afd091cf7d-task0-0.yc6e593b92ea46fd9ee188afd091cf7d.d37c68f328b94e8da56bde693916a55e.svc.cluster.local on Linux v3.10.0-1160.el7.x86_64 amd64 ```

10 months ago