I am thinking of creating a machine learning environment using Docker.
So I'm trying to build an environment using docker, but I can't run bash well.
After running docker-compose up
,
docker-compose exec ml/bin/bash
causes
ERROR—Displays No container found for ml_1
.
after running docker-compose up
or docker-compose up-d
docker ps-a
to check the container state, and it is exited
.
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
d67804c1bd66ml"/opt/nvidia/nvidia_…"9 minutes ago Exited(0)9 minutes ago desktop-ml-1
Additional
I checked and found that adding tty:true
to docker-compose.yml won't stop the container, so I did it in addition to docker-compose.yml.
ERROR:
Can't find a sustainable configuration file in this directory or any
parent.Are you in the right directory?
Supported filenames: docker-compose.yml, docker-compose.yaml
appears and does not work.
What should I do?The host environment is CentOS.
Dockerfile
FROM nvcr.io/nvidia/pytorch:21.10-py3
WORKDIR/home/workspace
# COPY./home/workspace
COPY requirements.txt${PWD}
RUN pip install-r requirements.txt
docker-compose.yml
version: '2.3'
services:
ml:
build:.
image —ml
volumes:
- .:/home/workspace
command —bash
deploy:
resources:
reservation:
devices:
- driver —nvidia
count:1
capabilities: [gpu]
tty —true
Without docker-compose, -it
was missing and could not enter bash.
The docker-compose method was resolved by adding tty:true
to docker-compose.yml.
ERROR:
Can't find a sustainable configuration file in this directory or any
parent.Are you in the right directory?
Supported filenames: docker-compose.yml, docker-compose.yaml
The problem in was the wrong directory to run docker-compose.
Thank you.
© 2024 OneMinuteCode. All rights reserved.