Neuro Platform API client
Project description
Table of Contents
- Preface
- neuro
- neuro admin
- neuro job
- neuro project
- neuro storage
- neuro image
- neuro config
- neuro completion
- neuro acl
- neuro help
- neuro run
- neuro submit
- neuro ps
- neuro status
- neuro exec
- neuro port-forward
- neuro logs
- neuro kill
- neuro top
- neuro save
- neuro login
- neuro logout
- neuro cp
- neuro ls
- neuro rm
- neuro mkdir
- neuro mv
- neuro images
- neuro push
- neuro pull
- neuro share
- Api
- Contributing
Preface
Welcome to Neuromation API Python client. Package ship command line tool called neuro. With neuro you can:
- Execute and debug jobs
- Manipulate Data
- Make some fun
neuro
Usage:
neuro [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
-v, --verbose | Give more output. Option is additive, and can be used up to 2 times. |
-q, --quiet | Give less output. Option is additive, and can be used up to 2 times. |
--neuromation-config PATH | Path to config directory. |
--show-traceback | Show python traceback on error, useful for debugging the tool. |
--color [yes | no | auto] | Color mode. |
--disable-pypi-version-check | Don't periodically check PyPI to determine whether a new version of Neuro Platform CLI is available for download. |
--network-timeout FLOAT | Network read timeout, seconds. |
--version | Show the version and exit. |
--trace | Trace sent HTTP requests and received replies to stderr. |
--hide-token / --no-hide-token | Prevent user's token sent in HTTP headers from being printed out to stderr during HTTP tracing. Can be used only together with option '--trace'. On by default. |
--help | Show this message and exit. |
Command Groups:
Usage | Description |
---|---|
neuro admin | Cluster administration commands |
neuro job | Job operations |
neuro project | Project operations |
neuro storage | Storage operations |
neuro image | Container image operations |
neuro config | Client configuration |
neuro completion | Output shell completion code |
neuro acl | Access Control List management |
Commands:
Usage | Description |
---|---|
neuro help | Get help on a command |
neuro run | Run a job with predefined resources configuration |
neuro submit | Submit an image to run on the cluster |
neuro ps | List all jobs |
neuro status | Display status of a job |
neuro exec | Execute command in a running job |
neuro port-forward | Forward port(s) of a running job to local port(s) |
neuro logs | Print the logs for a container |
neuro kill | Kill job(s) |
neuro top | Display GPU/CPU/Memory usage |
neuro save | Save job's state to an image |
neuro login | Log into Neuro Platform |
neuro logout | Log out |
neuro cp | Copy files and directories |
neuro ls | List directory contents |
neuro rm | Remove files or directories |
neuro mkdir | Make directories |
neuro mv | Move or rename files and directories |
neuro images | List images |
neuro push | Push an image to platform registry |
neuro pull | Pull an image from platform registry |
neuro share | Shares resource with another user |
neuro admin
Cluster administration commands.
Usage:
neuro admin [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro admin get-clusters | Print the list of available clusters |
neuro admin generate-cluster-config | Create a cluster configuration file |
neuro admin add-cluster | Create a new cluster and start its provisioning |
neuro admin get-cluster-users | Print the list of all users in the cluster with their assigned role |
neuro admin add-cluster-user | Add user access to specified cluster |
neuro admin remove-cluster-user | Remove user access from the cluster |
neuro admin set-user-quota | Set user quota to given values |
neuro admin add-user-quota | Add given values to user quota |
neuro admin get-clusters
Print the list of available clusters.
Usage:
neuro admin get-clusters [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro admin generate-cluster-config
Create a cluster configuration file.
Usage:
neuro admin generate-cluster-config [OPTIONS] [CONFIG]
Options:
Name | Description |
---|---|
--type [aws | gcp] | |
--help | Show this message and exit. |
neuro admin add-cluster
Create a new cluster and start its provisioning.
Usage:
neuro admin add-cluster [OPTIONS] CLUSTER_NAME CONFIG
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro admin get-cluster-users
Print the list of all users in the cluster with their assigned role.
Usage:
neuro admin get-cluster-users [OPTIONS] [CLUSTER_NAME]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro admin add-cluster-user
Add user access to specified cluster.
The command supports one of 3 user roles: admin, manager or user.
Usage:
neuro admin add-cluster-user [OPTIONS] CLUSTER_NAME USER_NAME [ROLE]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro admin remove-cluster-user
Remove user access from the cluster.
Usage:
neuro admin remove-cluster-user [OPTIONS] CLUSTER_NAME USER_NAME
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro admin set-user-quota
Set user quota to given values
Usage:
neuro admin set-user-quota [OPTIONS] CLUSTER_NAME USER_NAME
Options:
Name | Description |
---|---|
-g, --gpu AMOUNT | GPU quota value in hours (h) or minutes (m). |
-n, --non-gpu AMOUNT | Non-GPU quota value in hours (h) or minutes (m). |
--help | Show this message and exit. |
neuro admin add-user-quota
Add given values to user quota
Usage:
neuro admin add-user-quota [OPTIONS] CLUSTER_NAME USER_NAME
Options:
Name | Description |
---|---|
-g, --gpu AMOUNT | Additional GPU quota value in hours (h) or minutes (m). |
-n, --non-gpu AMOUNT | Additional non-GPU quota value in hours (h) or minutes (m). |
--help | Show this message and exit. |
neuro job
Job operations.
Usage:
neuro job [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro job run | Run a job with predefined resources configuration |
neuro job submit | Submit an image to run on the cluster |
neuro job ls | List all jobs |
neuro job status | Display status of a job |
neuro job exec | Execute command in a running job |
neuro job port-forward | Forward port(s) of a running job to local port(s) |
neuro job logs | Print the logs for a container |
neuro job kill | Kill job(s) |
neuro job top | Display GPU/CPU/Memory usage |
neuro job save | Save job's state to an image |
neuro job browse | Opens a job's URL in a web browser |
neuro job run
Run a job with predefined resources configuration.
IMAGE container image name.
CMD list will be passed as commands to model container.
Usage:
neuro job run [OPTIONS] IMAGE [CMD]...
Examples:
# Starts a container pytorch:latest on a machine with smaller GPU resources
# (see exact values in `neuro config show`) and with two volumes mounted:
# storage://<home-directory> --> /var/storage/home (in read-write mode),
# storage://neuromation/public --> /var/storage/neuromation (in read-only mode).
neuro run --preset=gpu-small --volume=HOME pytorch:latest
# Starts a container using the custom image my-ubuntu:latest stored in neuromation
# registry, run /script.sh and pass arg1 and arg2 as its arguments:
neuro run -s cpu-small image://~/my-ubuntu:latest --entrypoint=/script.sh arg1 arg2
Options:
Name | Description |
---|---|
-s, --preset PRESET | Predefined resource configuration (to see available values, run neuro config show ) |
-x, --extshm / -X, --no-extshm | Request extended '/dev/shm' space [default: True] |
--http PORT | Enable HTTP port forwarding to container [default: 80] |
--http-auth / --no-http-auth | Enable HTTP authentication for forwarded HTTP port [default: True] |
-n, --name NAME | Optional job name |
-d, --description DESC | Optional job description in free format |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-v, --volume MOUNT | Mounts directory from vault into container. Use multiple options to mount more than one volume. --volume=HOME is an alias for storage://~:/var/storage/home:rw and storage://neuromation/public:/var/storage/neuromation:ro |
--entrypoint TEXT | Executable entrypoint in the container (note that it overwrites ENTRYPOINT and CMD instructions of the docker image) |
-e, --env VAR=VAL | Set environment variable in container Use multiple options to define more than one variable |
--env-file PATH | File with environment variables to pass |
--wait-start / --no-wait-start | Wait for a job start or failure [default: True] |
--pass-config / --no-pass-config | Upload neuro config to the job [default: False] |
--browse | Open a job's URL in a web browser |
--detach | Don't attach to job logs and don't wait for exit code |
--help | Show this message and exit. |
neuro job submit
Submit an image to run on the cluster.
IMAGE container image name.
CMD list will be passed as commands to model container.
Usage:
neuro job submit [OPTIONS] IMAGE [CMD]...
Examples:
# Starts a container pytorch:latest with two paths mounted. Directory /q1/
# is mounted in read only mode to /qm directory within container.
# Directory /mod mounted to /mod directory in read-write mode.
neuro submit --volume storage:/q1:/qm:ro --volume storage:/mod:/mod:rw pytorch:latest
# Starts a container using the custom image my-ubuntu:latest stored in neuromation
# registry, run /script.sh and pass arg1 arg2 arg3 as its arguments:
neuro submit image://~/my-ubuntu:latest --entrypoint=/script.sh arg1 arg2 arg3
Options:
Name | Description |
---|---|
-g, --gpu NUMBER | Number of GPUs to request [default: 0] |
--gpu-model MODEL | GPU to use [default: nvidia-tesla-k80] |
--tpu-type TYPE | TPU type to use |
--tpu-sw-version VERSION | Requested TPU software version |
-c, --cpu NUMBER | Number of CPUs to request [default: 0.1] |
-m, --memory AMOUNT | Memory amount to request [default: 1G] |
-x, --extshm / -X, --no-extshm | Request extended '/dev/shm' space [default: True] |
--http PORT | Enable HTTP port forwarding to container |
--http-auth / --no-http-auth | Enable HTTP authentication for forwarded HTTP port [default: True] |
-p, --preemptible / -P, --non-preemptible | Run job on a lower-cost preemptible instance [default: False] |
-n, --name NAME | Optional job name |
-d, --description DESC | Optional job description in free format |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-v, --volume MOUNT | Mounts directory from vault into container. Use multiple options to mount more than one volume. --volume=HOME is an alias for storage://~:/var/storage/home:rw and storage://neuromation/public:/var/storage/neuromation:ro |
--entrypoint TEXT | Executable entrypoint in the container (note that it overwrites ENTRYPOINT and CMD instructions of the docker image) |
-e, --env VAR=VAL | Set environment variable in container Use multiple options to define more than one variable |
--env-file PATH | File with environment variables to pass |
--wait-start / --no-wait-start | Wait for a job start or failure [default: True] |
--pass-config / --no-pass-config | Upload neuro config to the job [default: False] |
--browse | Open a job's URL in a web browser |
--detach | Don't attach to job logs and don't wait for exit code |
--help | Show this message and exit. |
neuro job ls
List all jobs.
Usage:
neuro job ls [OPTIONS]
Examples:
neuro ps -a
neuro ps -a --owner=user-1 --owner=user-2
neuro ps --name my-experiments-v1 -s failed -s succeeded
neuro ps --description=my favourite job
neuro ps -s failed -s succeeded -q
Options:
Name | Description |
---|---|
-s, --status [pending | running | succeeded | failed | all] | Filter out jobs by status (multiple option). Note: option all is deprecated, use neuro ps -a instead. |
-o, --owner TEXT | Filter out jobs by owner (multiple option). |
-a, --all | Show all jobs regardless the status (equivalent to \-s pending -s running -s succeeded -s failed ). |
-n, --name NAME | Filter out jobs by name. |
-d, --description DESCRIPTION | Filter out jobs by description (exact match). |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-w, --wide | Do not cut long lines for terminal width. |
--format COLUMNS | Output table format, see "neuro help ps-format" for more info about the format specification. The default can be changed using the job.ps-format configuration variable documented in "neuro help user-config" |
--help | Show this message and exit. |
neuro job status
Display status of a job.
Usage:
neuro job status [OPTIONS] JOB
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro job exec
Execute command in a running job.
Usage:
neuro job exec [OPTIONS] JOB CMD...
Examples:
# Provides a shell to the container:
neuro exec my-job /bin/bash
# Executes a single command in the container and returns the control:
neuro exec --no-tty my-job ls -l
Options:
Name | Description |
---|---|
-t, --tty / -T, --no-tty | Allocate virtual tty. Useful for interactive jobs. |
--no-key-check | Disable host key checks. Should be used with caution. |
--timeout FLOAT | Maximum allowed time for executing the command, 0 for no timeout [default: 0] |
--help | Show this message and exit. |
neuro job port-forward
Forward port(s) of a running job to local port(s).
Usage:
neuro job port-forward [OPTIONS] JOB LOCAL_REMOTE_PORT...
Examples:
# Forward local port 2080 to port 80 of job's container.
# You can use http://localhost:2080 in browser to access job's served http
neuro job port-forward my-fastai-job 2080:80
# Forward local port 2222 to job's port 22
# Then copy all data from container's folder '/data' to current folder
# (please run second command in other terminal)
neuro job port-forward my-job-with-ssh-server 2222:22
rsync -avxzhe ssh -p 2222 root@localhost:/data .
# Forward few ports at once
neuro job port-forward my-job- 2080:80 2222:22 2000:100
Options:
Name | Description |
---|---|
--no-key-check | Disable host key checks. Should be used with caution. |
--help | Show this message and exit. |
neuro job logs
Print the logs for a container.
Usage:
neuro job logs [OPTIONS] JOB
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro job kill
Kill job(s).
Usage:
neuro job kill [OPTIONS] JOBS...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro job top
Display GPU/CPU/Memory usage.
Usage:
neuro job top [OPTIONS] JOB
Options:
Name | Description |
---|---|
--timeout FLOAT | Maximum allowed time for executing the command, 0 for no timeout [default: 0] |
--help | Show this message and exit. |
neuro job save
Save job's state to an image.
Usage:
neuro job save [OPTIONS] JOB IMAGE
Examples:
neuro job save job-id image:ubuntu-patched
neuro job save my-favourite-job image://~/ubuntu-patched:v1
neuro job save my-favourite-job image://bob/ubuntu-patched
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro job browse
Opens a job's URL in a web browser.
Usage:
neuro job browse [OPTIONS] JOB
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro project
Project operations.
Usage:
neuro project [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro project init | Initialize an empty project |
neuro project init
Initialize an empty project.
Usage:
neuro project init [OPTIONS] [SLUG]
Examples:
# Initializes a scaffolding for the new project with the recommended project
# structure (see http://github.com/neuromation/cookiecutter-neuro-project)
neuro project init
# Initializes a scaffolding for the new project with the recommended project
# structure and sets default project folder name to "example"
neuro project init my-project-id
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro storage
Storage operations.
Usage:
neuro storage [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro storage cp | Copy files and directories |
neuro storage ls | List directory contents |
neuro storage glob | List resources that match PATTERNS |
neuro storage rm | Remove files or directories |
neuro storage mkdir | Make directories |
neuro storage mv | Move or rename files and directories |
neuro storage load | Copy files and directories using MinIO (EXPERIMENTAL) |
neuro storage cp
Copy files and directories.
Either SOURCES or DESTINATION should have storage:// scheme. If scheme is
omitted, file:// scheme is assumed.
Use /dev/stdin and /dev/stdout file names to copy a file from terminal and
print the content of file on the storage to console.
Usage:
neuro storage cp [OPTIONS] [SOURCES]... [DESTINATION]
Examples:
# copy local files into remote storage root
neuro cp foo.txt bar/baz.dat storage:
neuro cp foo.txt bar/baz.dat -t storage:
# copy local directory `foo` into existing remote directory `bar`
neuro cp -r foo -t storage:bar
# copy the content of local directory `foo` into existing remote
# directory `bar`
neuro cp -r -T storage:foo storage:bar
# download remote file `foo.txt` into local file `/tmp/foo.txt` with
# explicit file:// scheme set
neuro cp storage:foo.txt file:///tmp/foo.txt
neuro cp -T storage:foo.txt file:///tmp/foo.txt
neuro cp storage:foo.txt file:///tmp
neuro cp storage:foo.txt -t file:///tmp
# download other user's remote file into the current directory
neuro cp storage://{username}/foo.txt .
# download only files with extension `.out` into the current directory
neuro cp storage:results/*.out .
Options:
Name | Description |
---|---|
-r, --recursive | Recursive copy, off by default |
--glob / --no-glob | Expand glob patterns in SOURCES with explicit scheme. [default: True] |
-t, --target-directory DIRECTORY | Copy all SOURCES into DIRECTORY. |
-T, --no-target-directory | Treat DESTINATION as a normal file. |
-u, --update | Copy only when the SOURCE file is newer than the destination file or when the destination file is missing. |
--exclude | Exclude files and directories that match the specified pattern. The default can be changed using the storage.cp-exclude configuration variable documented in "neuro help user-config" |
--include | Don't exclude files and directories that match the specified pattern. The default can be changed using the storage.cp-exclude configuration variable documented in "neuro help user-config" |
-p, --progress / -P, --no-progress | Show progress, on by default. |
--help | Show this message and exit. |
neuro storage ls
List directory contents.
By default PATH is equal user's home dir (storage:)
Usage:
neuro storage ls [OPTIONS] [PATHS]...
Options:
Name | Description |
---|---|
-h, --human-readable | with -l print human readable sizes (e.g., 2K, 540M) |
-l | use a long listing format |
--sort [name | size | time] | sort by given field, default is name |
-d, --directory | list directories themselves, not their contents |
--help | Show this message and exit. |
neuro storage glob
List resources that match PATTERNS.
Usage:
neuro storage glob [OPTIONS] [PATTERNS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro storage rm
Remove files or directories.
Usage:
neuro storage rm [OPTIONS] PATHS...
Examples:
neuro rm storage:foo/bar
neuro rm storage://{username}/foo/bar
neuro rm --recursive storage://{username}/foo/
neuro rm storage:foo/**/*.tmp
Options:
Name | Description |
---|---|
-r, --recursive | remove directories and their contents recursively |
--glob / --no-glob | Expand glob patterns in PATHS [default: True] |
--help | Show this message and exit. |
neuro storage mkdir
Make directories.
Usage:
neuro storage mkdir [OPTIONS] PATHS...
Options:
Name | Description |
---|---|
-p, --parents | No error if existing, make parent directories as needed |
--help | Show this message and exit. |
neuro storage mv
Move or rename files and directories.
SOURCE must contain path to the file or directory existing on the storage,
and DESTINATION must contain the full path to the target file or directory.
Usage:
neuro storage mv [OPTIONS] [SOURCES]... [DESTINATION]
Examples:
# move and rename remote file
neuro mv storage:foo.txt storage:bar/baz.dat
neuro mv -T storage:foo.txt storage:bar/baz.dat
# move remote files into existing remote directory
neuro mv storage:foo.txt storage:bar/baz.dat storage:dst
neuro mv storage:foo.txt storage:bar/baz.dat -t storage:dst
# move the content of remote directory into other existing
# remote directory
neuro mv -T storage:foo storage:bar
# move remote file into other user's directory
neuro mv storage:foo.txt storage://{username}/bar.dat
# move remote file from other user's directory
neuro mv storage://{username}/foo.txt storage:bar.dat
Options:
Name | Description |
---|---|
--glob / --no-glob | Expand glob patterns in SOURCES [default: True] |
-t, --target-directory DIRECTORY | Copy all SOURCES into DIRECTORY |
-T, --no-target-directory | Treat DESTINATION as a normal file |
--help | Show this message and exit. |
neuro storage load
Copy files and directories using MinIO (EXPERIMENTAL).
Same as "cp", but uses MinIO and the Amazon S3 protocol.
(DEPRECATED)
Usage:
neuro storage load [OPTIONS] [SOURCES]... [DESTINATION]
Options:
Name | Description |
---|---|
-r, --recursive | Recursive copy, off by default |
--glob / --no-glob | Expand glob patterns in SOURCES with explicit scheme [default: True] |
-t, --target-directory DIRECTORY | Copy all SOURCES into DIRECTORY |
-T, --no-target-directory | Treat DESTINATION as a normal file |
-u, --update | Copy only when the SOURCE file is newer than the destination file or when the destination file is missing |
-p, --progress | Show progress, off by default |
--help | Show this message and exit. |
neuro image
Container image operations.
Usage:
neuro image [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro image ls | List images |
neuro image push | Push an image to platform registry |
neuro image pull | Pull an image from platform registry |
neuro image tags | List tags for image in platform registry |
neuro image ls
List images.
Usage:
neuro image ls [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro image push
Push an image to platform registry.
Remote image must be URL with image:// scheme. Image names can contain tag.
If tags not specified 'latest' will be used as value.
Usage:
neuro image push [OPTIONS] LOCAL_IMAGE [REMOTE_IMAGE]
Examples:
neuro push myimage
neuro push alpine:latest image:my-alpine:production
neuro push alpine image://myfriend/alpine:shared
Options:
Name | Description |
---|---|
-q, --quiet | Run command in quiet mode (DEPRECATED) |
--help | Show this message and exit. |
neuro image pull
Pull an image from platform registry.
Remote image name must be URL with image:// scheme. Image names can contain
tag.
Usage:
neuro image pull [OPTIONS] REMOTE_IMAGE [LOCAL_IMAGE]
Examples:
neuro pull image:myimage
neuro pull image://myfriend/alpine:shared
neuro pull image://username/my-alpine:production alpine:from-registry
Options:
Name | Description |
---|---|
-q, --quiet | Run command in quiet mode (DEPRECATED) |
--help | Show this message and exit. |
neuro image tags
List tags for image in platform registry.
Image name must be URL with image:// scheme.
Usage:
neuro image tags [OPTIONS] IMAGE
Examples:
neuro image tags image://myfriend/alpine
neuro image tags image:myimage
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config
Client configuration.
Usage:
neuro config [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro config login | Log into Neuro Platform |
neuro config login-with-token | Log into Neuro Platform with token |
neuro config login-headless | Log into Neuro Platform from non-GUI server environment |
neuro config show | Print current settings |
neuro config show-token | Print current authorization token |
neuro config show-quota | Print quota and remaining computation time for active cluster |
neuro config get-clusters | Fetch and display the list of available clusters |
neuro config switch-cluster | Switch the active cluster |
neuro config docker | Configure docker client to fit the Neuro Platform |
neuro config logout | Log out |
neuro config login
Log into Neuro Platform.
URL is a platform entrypoint URL.
Usage:
neuro config login [OPTIONS] [URL]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config login-with-token
Log into Neuro Platform with token.
TOKEN is authentication token provided by administration team. URL is a
platform entrypoint URL.
Usage:
neuro config login-with-token [OPTIONS] TOKEN [URL]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config login-headless
Log into Neuro Platform from non-GUI server environment.
URL is a platform entrypoint URL.
The command works similar to "neuro login" but instead of opening a browser
for performing OAuth registration prints an URL that should be open on guest
host.
Then user inputs a code displayed in a browser after successful login back
in neuro command to finish the login process.
Usage:
neuro config login-headless [OPTIONS] [URL]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config show
Print current settings.
Usage:
neuro config show [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config show-token
Print current authorization token.
Usage:
neuro config show-token [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config show-quota
Print quota and remaining computation time for active cluster.
Usage:
neuro config show-quota [OPTIONS] [USER]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config get-clusters
Fetch and display the list of available clusters.
Usage:
neuro config get-clusters [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config switch-cluster
Switch the active cluster.
CLUSTER_NAME is the cluster name to select. The interactive prompt is used
if the name is omitted (default).
Usage:
neuro config switch-cluster [OPTIONS] [CLUSTER_NAME]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro config docker
Configure docker client to fit the Neuro Platform.
Usage:
neuro config docker [OPTIONS]
Options:
Name | Description |
---|---|
--docker-config PATH | Specifies the location of the Docker client configuration files |
--help | Show this message and exit. |
neuro config logout
Log out.
Usage:
neuro config logout [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro completion
Output shell completion code.
Usage:
neuro completion [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro completion generate | Provide an instruction for shell completion generation |
neuro completion patch | Automatically patch shell configuration profile to enable completion |
neuro completion generate
Provide an instruction for shell completion generation.
Usage:
neuro completion generate [OPTIONS] [bash|zsh]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro completion patch
Automatically patch shell configuration profile to enable completion
Usage:
neuro completion patch [OPTIONS] [bash|zsh]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro acl
Access Control List management.
Usage:
neuro acl [OPTIONS] COMMAND [ARGS]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Commands:
Usage | Description |
---|---|
neuro acl grant | Shares resource with another user |
neuro acl revoke | Revoke user access from another user |
neuro acl list | List shared resources |
neuro acl grant
Shares resource with another user.
URI shared resource.
USER username to share resource with.
PERMISSION sharing access right: read, write, or manage.
Usage:
neuro acl grant [OPTIONS] URI USER [read|write|manage]
Examples:
neuro acl grant storage:///sample_data/ alice manage
neuro acl grant image:resnet50 bob read
neuro acl grant job:///my_job_id alice write
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro acl revoke
Revoke user access from another user.
URI previously shared resource to revoke.
USER to revoke URI resource from.
Usage:
neuro acl revoke [OPTIONS] URI USER
Examples:
neuro acl revoke storage:///sample_data/ alice
neuro acl revoke image:resnet50 bob
neuro acl revoke job:///my_job_id alice
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro acl list
List shared resources.
The command displays a list of resources shared BY current user (default).
To display a list of resources shared WITH current user apply --shared
option.
Usage:
neuro acl list [OPTIONS]
Examples:
neuro acl list
neuro acl list --scheme storage
neuro acl list --shared
neuro acl list --shared --scheme image
Options:
Name | Description |
---|---|
-s, --scheme TEXT | Filter resources by scheme, e.g. job, storage, image or user. |
--shared | Output the resources shared by the user. |
--help | Show this message and exit. |
neuro help
Get help on a command.
Usage:
neuro help [OPTIONS] [COMMAND]...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro run
Run a job with predefined resources configuration.
IMAGE container image name.
CMD list will be passed as commands to model container.
Usage:
neuro run [OPTIONS] IMAGE [CMD]...
Examples:
# Starts a container pytorch:latest on a machine with smaller GPU resources
# (see exact values in `neuro config show`) and with two volumes mounted:
# storage://<home-directory> --> /var/storage/home (in read-write mode),
# storage://neuromation/public --> /var/storage/neuromation (in read-only mode).
neuro run --preset=gpu-small --volume=HOME pytorch:latest
# Starts a container using the custom image my-ubuntu:latest stored in neuromation
# registry, run /script.sh and pass arg1 and arg2 as its arguments:
neuro run -s cpu-small image://~/my-ubuntu:latest --entrypoint=/script.sh arg1 arg2
Options:
Name | Description |
---|---|
-s, --preset PRESET | Predefined resource configuration (to see available values, run neuro config show ) |
-x, --extshm / -X, --no-extshm | Request extended '/dev/shm' space [default: True] |
--http PORT | Enable HTTP port forwarding to container [default: 80] |
--http-auth / --no-http-auth | Enable HTTP authentication for forwarded HTTP port [default: True] |
-n, --name NAME | Optional job name |
-d, --description DESC | Optional job description in free format |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-v, --volume MOUNT | Mounts directory from vault into container. Use multiple options to mount more than one volume. --volume=HOME is an alias for storage://~:/var/storage/home:rw and storage://neuromation/public:/var/storage/neuromation:ro |
--entrypoint TEXT | Executable entrypoint in the container (note that it overwrites ENTRYPOINT and CMD instructions of the docker image) |
-e, --env VAR=VAL | Set environment variable in container Use multiple options to define more than one variable |
--env-file PATH | File with environment variables to pass |
--wait-start / --no-wait-start | Wait for a job start or failure [default: True] |
--pass-config / --no-pass-config | Upload neuro config to the job [default: False] |
--browse | Open a job's URL in a web browser |
--detach | Don't attach to job logs and don't wait for exit code |
--help | Show this message and exit. |
neuro submit
Submit an image to run on the cluster.
IMAGE container image name.
CMD list will be passed as commands to model container.
Usage:
neuro submit [OPTIONS] IMAGE [CMD]...
Examples:
# Starts a container pytorch:latest with two paths mounted. Directory /q1/
# is mounted in read only mode to /qm directory within container.
# Directory /mod mounted to /mod directory in read-write mode.
neuro submit --volume storage:/q1:/qm:ro --volume storage:/mod:/mod:rw pytorch:latest
# Starts a container using the custom image my-ubuntu:latest stored in neuromation
# registry, run /script.sh and pass arg1 arg2 arg3 as its arguments:
neuro submit image://~/my-ubuntu:latest --entrypoint=/script.sh arg1 arg2 arg3
Options:
Name | Description |
---|---|
-g, --gpu NUMBER | Number of GPUs to request [default: 0] |
--gpu-model MODEL | GPU to use [default: nvidia-tesla-k80] |
--tpu-type TYPE | TPU type to use |
--tpu-sw-version VERSION | Requested TPU software version |
-c, --cpu NUMBER | Number of CPUs to request [default: 0.1] |
-m, --memory AMOUNT | Memory amount to request [default: 1G] |
-x, --extshm / -X, --no-extshm | Request extended '/dev/shm' space [default: True] |
--http PORT | Enable HTTP port forwarding to container |
--http-auth / --no-http-auth | Enable HTTP authentication for forwarded HTTP port [default: True] |
-p, --preemptible / -P, --non-preemptible | Run job on a lower-cost preemptible instance [default: False] |
-n, --name NAME | Optional job name |
-d, --description DESC | Optional job description in free format |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-v, --volume MOUNT | Mounts directory from vault into container. Use multiple options to mount more than one volume. --volume=HOME is an alias for storage://~:/var/storage/home:rw and storage://neuromation/public:/var/storage/neuromation:ro |
--entrypoint TEXT | Executable entrypoint in the container (note that it overwrites ENTRYPOINT and CMD instructions of the docker image) |
-e, --env VAR=VAL | Set environment variable in container Use multiple options to define more than one variable |
--env-file PATH | File with environment variables to pass |
--wait-start / --no-wait-start | Wait for a job start or failure [default: True] |
--pass-config / --no-pass-config | Upload neuro config to the job [default: False] |
--browse | Open a job's URL in a web browser |
--detach | Don't attach to job logs and don't wait for exit code |
--help | Show this message and exit. |
neuro ps
List all jobs.
Usage:
neuro ps [OPTIONS]
Examples:
neuro ps -a
neuro ps -a --owner=user-1 --owner=user-2
neuro ps --name my-experiments-v1 -s failed -s succeeded
neuro ps --description=my favourite job
neuro ps -s failed -s succeeded -q
Options:
Name | Description |
---|---|
-s, --status [pending | running | succeeded | failed | all] | Filter out jobs by status (multiple option). Note: option all is deprecated, use neuro ps -a instead. |
-o, --owner TEXT | Filter out jobs by owner (multiple option). |
-a, --all | Show all jobs regardless the status (equivalent to \-s pending -s running -s succeeded -s failed ). |
-n, --name NAME | Filter out jobs by name. |
-d, --description DESCRIPTION | Filter out jobs by description (exact match). |
-q, --quiet | Run command in quiet mode (DEPRECATED) |
-w, --wide | Do not cut long lines for terminal width. |
--format COLUMNS | Output table format, see "neuro help ps-format" for more info about the format specification. The default can be changed using the job.ps-format configuration variable documented in "neuro help user-config" |
--help | Show this message and exit. |
neuro status
Display status of a job.
Usage:
neuro status [OPTIONS] JOB
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro exec
Execute command in a running job.
Usage:
neuro exec [OPTIONS] JOB CMD...
Examples:
# Provides a shell to the container:
neuro exec my-job /bin/bash
# Executes a single command in the container and returns the control:
neuro exec --no-tty my-job ls -l
Options:
Name | Description |
---|---|
-t, --tty / -T, --no-tty | Allocate virtual tty. Useful for interactive jobs. |
--no-key-check | Disable host key checks. Should be used with caution. |
--timeout FLOAT | Maximum allowed time for executing the command, 0 for no timeout [default: 0] |
--help | Show this message and exit. |
neuro port-forward
Forward port(s) of a running job to local port(s).
Usage:
neuro port-forward [OPTIONS] JOB LOCAL_REMOTE_PORT...
Examples:
# Forward local port 2080 to port 80 of job's container.
# You can use http://localhost:2080 in browser to access job's served http
neuro job port-forward my-fastai-job 2080:80
# Forward local port 2222 to job's port 22
# Then copy all data from container's folder '/data' to current folder
# (please run second command in other terminal)
neuro job port-forward my-job-with-ssh-server 2222:22
rsync -avxzhe ssh -p 2222 root@localhost:/data .
# Forward few ports at once
neuro job port-forward my-job- 2080:80 2222:22 2000:100
Options:
Name | Description |
---|---|
--no-key-check | Disable host key checks. Should be used with caution. |
--help | Show this message and exit. |
neuro logs
Print the logs for a container.
Usage:
neuro logs [OPTIONS] JOB
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro kill
Kill job(s).
Usage:
neuro kill [OPTIONS] JOBS...
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro top
Display GPU/CPU/Memory usage.
Usage:
neuro top [OPTIONS] JOB
Options:
Name | Description |
---|---|
--timeout FLOAT | Maximum allowed time for executing the command, 0 for no timeout [default: 0] |
--help | Show this message and exit. |
neuro save
Save job's state to an image.
Usage:
neuro save [OPTIONS] JOB IMAGE
Examples:
neuro job save job-id image:ubuntu-patched
neuro job save my-favourite-job image://~/ubuntu-patched:v1
neuro job save my-favourite-job image://bob/ubuntu-patched
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro login
Log into Neuro Platform.
URL is a platform entrypoint URL.
Usage:
neuro login [OPTIONS] [URL]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro logout
Log out.
Usage:
neuro logout [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro cp
Copy files and directories.
Either SOURCES or DESTINATION should have storage:// scheme. If scheme is
omitted, file:// scheme is assumed.
Use /dev/stdin and /dev/stdout file names to copy a file from terminal and
print the content of file on the storage to console.
Usage:
neuro cp [OPTIONS] [SOURCES]... [DESTINATION]
Examples:
# copy local files into remote storage root
neuro cp foo.txt bar/baz.dat storage:
neuro cp foo.txt bar/baz.dat -t storage:
# copy local directory `foo` into existing remote directory `bar`
neuro cp -r foo -t storage:bar
# copy the content of local directory `foo` into existing remote
# directory `bar`
neuro cp -r -T storage:foo storage:bar
# download remote file `foo.txt` into local file `/tmp/foo.txt` with
# explicit file:// scheme set
neuro cp storage:foo.txt file:///tmp/foo.txt
neuro cp -T storage:foo.txt file:///tmp/foo.txt
neuro cp storage:foo.txt file:///tmp
neuro cp storage:foo.txt -t file:///tmp
# download other user's remote file into the current directory
neuro cp storage://{username}/foo.txt .
# download only files with extension `.out` into the current directory
neuro cp storage:results/*.out .
Options:
Name | Description |
---|---|
-r, --recursive | Recursive copy, off by default |
--glob / --no-glob | Expand glob patterns in SOURCES with explicit scheme. [default: True] |
-t, --target-directory DIRECTORY | Copy all SOURCES into DIRECTORY. |
-T, --no-target-directory | Treat DESTINATION as a normal file. |
-u, --update | Copy only when the SOURCE file is newer than the destination file or when the destination file is missing. |
--exclude | Exclude files and directories that match the specified pattern. The default can be changed using the storage.cp-exclude configuration variable documented in "neuro help user-config" |
--include | Don't exclude files and directories that match the specified pattern. The default can be changed using the storage.cp-exclude configuration variable documented in "neuro help user-config" |
-p, --progress / -P, --no-progress | Show progress, on by default. |
--help | Show this message and exit. |
neuro ls
List directory contents.
By default PATH is equal user's home dir (storage:)
Usage:
neuro ls [OPTIONS] [PATHS]...
Options:
Name | Description |
---|---|
-h, --human-readable | with -l print human readable sizes (e.g., 2K, 540M) |
-l | use a long listing format |
--sort [name | size | time] | sort by given field, default is name |
-d, --directory | list directories themselves, not their contents |
--help | Show this message and exit. |
neuro rm
Remove files or directories.
Usage:
neuro rm [OPTIONS] PATHS...
Examples:
neuro rm storage:foo/bar
neuro rm storage://{username}/foo/bar
neuro rm --recursive storage://{username}/foo/
neuro rm storage:foo/**/*.tmp
Options:
Name | Description |
---|---|
-r, --recursive | remove directories and their contents recursively |
--glob / --no-glob | Expand glob patterns in PATHS [default: True] |
--help | Show this message and exit. |
neuro mkdir
Make directories.
Usage:
neuro mkdir [OPTIONS] PATHS...
Options:
Name | Description |
---|---|
-p, --parents | No error if existing, make parent directories as needed |
--help | Show this message and exit. |
neuro mv
Move or rename files and directories.
SOURCE must contain path to the file or directory existing on the storage,
and DESTINATION must contain the full path to the target file or directory.
Usage:
neuro mv [OPTIONS] [SOURCES]... [DESTINATION]
Examples:
# move and rename remote file
neuro mv storage:foo.txt storage:bar/baz.dat
neuro mv -T storage:foo.txt storage:bar/baz.dat
# move remote files into existing remote directory
neuro mv storage:foo.txt storage:bar/baz.dat storage:dst
neuro mv storage:foo.txt storage:bar/baz.dat -t storage:dst
# move the content of remote directory into other existing
# remote directory
neuro mv -T storage:foo storage:bar
# move remote file into other user's directory
neuro mv storage:foo.txt storage://{username}/bar.dat
# move remote file from other user's directory
neuro mv storage://{username}/foo.txt storage:bar.dat
Options:
Name | Description |
---|---|
--glob / --no-glob | Expand glob patterns in SOURCES [default: True] |
-t, --target-directory DIRECTORY | Copy all SOURCES into DIRECTORY |
-T, --no-target-directory | Treat DESTINATION as a normal file |
--help | Show this message and exit. |
neuro images
List images.
Usage:
neuro images [OPTIONS]
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
neuro push
Push an image to platform registry.
Remote image must be URL with image:// scheme. Image names can contain tag.
If tags not specified 'latest' will be used as value.
Usage:
neuro push [OPTIONS] LOCAL_IMAGE [REMOTE_IMAGE]
Examples:
neuro push myimage
neuro push alpine:latest image:my-alpine:production
neuro push alpine image://myfriend/alpine:shared
Options:
Name | Description |
---|---|
-q, --quiet | Run command in quiet mode (DEPRECATED) |
--help | Show this message and exit. |
neuro pull
Pull an image from platform registry.
Remote image name must be URL with image:// scheme. Image names can contain
tag.
Usage:
neuro pull [OPTIONS] REMOTE_IMAGE [LOCAL_IMAGE]
Examples:
neuro pull image:myimage
neuro pull image://myfriend/alpine:shared
neuro pull image://username/my-alpine:production alpine:from-registry
Options:
Name | Description |
---|---|
-q, --quiet | Run command in quiet mode (DEPRECATED) |
--help | Show this message and exit. |
neuro share
Shares resource with another user.
URI shared resource.
USER username to share resource with.
PERMISSION sharing access right: read, write, or manage.
Usage:
neuro share [OPTIONS] URI USER [read|write|manage]
Examples:
neuro acl grant storage:///sample_data/ alice manage
neuro acl grant image:resnet50 bob read
neuro acl grant job:///my_job_id alice write
Options:
Name | Description |
---|---|
--help | Show this message and exit. |
Api
TODO
Contributing
git clone https://github.com/neuromation/platform-api-clients.git
cd platform-api-clients/python
For OSX users install coreutils to properly interpret shell commands:
brew install coreutils
Before you begin, it is recommended to have clean virtual environment installed:
python -m venv .env
source .env/bin/activate
Development flow:
- Install dependencies:
make init
- Run tests:
make test
- Lint:
make lint
- Publish to pypi:
make publish
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for neuromation-20.1.15-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 95cb950cb47eae7d5710975a7db5971ff7126c65126211b1a2a4ff53119ddb38 |
|
MD5 | d9843fc06e7295ed88e5cbb14dc87b89 |
|
BLAKE2b-256 | 47eea2d336ed24fc69688f12608496e9eecd768759b1617f95e33b15c49f31af |