xinference安装报错 #674
Labels
No Label
bug
duplicate
enhancement
help wanted
invalid
question
wontfix
No Milestone
No project
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: HswOAuth/llm_course#674
Loading…
Reference in New Issue
Block a user
No description provided.
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
老师，我用windows安装报错如下，安装的非常慢，可以用windows安装吗？电脑内存12g,8核。
docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v $(pwd)/.xinference:D:\software\AItool\xinference -v $(pwd):/home -v $(pwd)/.cache/huggingface:/root/.cache/huggingface -v $(pwd)/.cache/modelscope:/root/.cache/modelscope -p 9997:9997 --gpus all registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference xinference-local -H 0.0.0.0
老师，这个命令是不是有问题？
docker run -d
-e XINFERENCE_MODEL_SRC=modelscope
-v /d/software/AItool/xinference:/root/.xinference
-v "$(pwd)":/home
-v "$(pwd)/.cache/huggingface":/root/.cache/huggingface
-v "$(pwd)/.cache/modelscope":/root/.cache/modelscope
-p 9997:9997
--gpus all
registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference
xinference-local -H 0.0.0.0 可以对照下这个命令
我按老师发的这种命令运行，每一行是一条命令，连不上，我就放到同一行了
老师，能往上翻翻看看吗？报错的
-v $(pwd)/.xinference:D:\software\AItool\xinference改成这个试试-v /d/software/AItool/xinference:/root/.xinference
或者-v $(pwd)/.xinference:/d/software/AItool/xinference
老师，这俩都报错。
老师，这个能看看吗？
你的电脑有gpu且安装了wsl2用这个：docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v ./.xinference:/root/.xinference -v ./.cache/huggingface:/root/.cache/huggingface -v ./.cache/modelscope:/root/.cache/modelscope -p 9997:9997 --gpus all registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference xinference-local -H 0.0.0.0
你的电脑没有gpu用这个：docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v ./.xinference:/root/.xinference -v ./.cache/huggingface:/root/.cache/huggingface -v ./.cache/modelscope:/root/.cache/modelscope -p 9997:9997 registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference xinference-local -H 0.0.0.0
老师，我把 .去了？
行，把.去掉吧
老师，还是不行
/cache/huggingface 改成./cache/huggingface
docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v ./xinference:/root/.xinference -v ./cache/huggingface:/root/.cache/huggingface -v ./cache/modelscope:/root/.cache/modelscope -p 9997:9997 registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference xinference-local -H 0.0.0.0
老师，我运行的代码 就是 ./cache/huggingface 带 . 的
docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v /d/software/AItool/xinference:/root/.xinference -v $PWD/.cache/huggingface:/root/.cache/huggingface -v $PWD/.cache/modelscope:/root/.cache/modelscope -p 9997:9997 registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference 试下这个
不行，无效的字符，识别不了PWD,
docker run -d -e XINFERENCE_MODEL_SRC=modelscope -v d:/software/AItool/xinference:/root/.xinference -v d:/software/AItool/xinference/.cache/huggingface:/root/.cache/huggingface -v d:/software/AItool/xinference/.cache/modelscope:/root/.cache/modelscope -p 9997:9997 registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference 试下这个
老师，输出一串数字
启动成功了，这串数字就是xinference的容器ID，可以访问127.0.0.1:9997进入xinference
好的，感谢老师！！！总算成功了
老师，docker desktop 里这俩启动不了，看不到报错信息，页面无法访问
我又把这俩都删了，重新启动了下，还是这样，启动后，秒失败
registry.cn-hangzhou.aliyuncs.com/xprobe_xinference/xinference:latest-cpu使用这个镜像
老师，这个命令下载完镜像，启动容器后，秒失败
老师，什么时候方便远程看一下呢？
老师，我在https://blog.csdn.net/bdbox/article/details/142391649 ， https://blog.csdn.net/u011788214/article/details/141255352?spm=1001.2101.3001.6650.3&utm_medium=distribute.pc_relevant.none-task-blog-2%7Edefault%7ECTRLIST%7ERate-3-141255352-blog-142391649.235%5Ev43%5Econtrol&depth_1-utm_source=distribute.pc_relevant.none-task-blog-2%7Edefault%7ECTRLIST%7ERate-3-141255352-blog-142391649.235%5Ev43%5Econtrol&utm_relevant_index=4
两个网页上，看到这个命令 xinference-local -H 0.0.0.0，我加上后，容器能启动成功了，但是http://127.0.0.1:9997,http://localhost:9997 提示无法访问， dockers desktop日志提示访问 http://0.0.0.0:9997 ,但也访问不了
这个是xinference的容器的问题，应该是没有完全适配好windows，就是还可以直接通过pip install安装xinference，pip install "xinference[transformers]"
xinference的参考安装地址：https://inference.readthedocs.io/zh-cn/latest/getting_started/installation.html
在电脑安linux 虚拟机，用这个命令或者租服务器？
这几种方式都可以，但是大模型的运行需要消耗比较大的资源，可以先租个性能好的服务器做练习