CLI Reference
tensorkube deploy
Deploy your containerized application on Tensorkube. This command requires a dockerfile to be present in the current directory.
Usage
Options
-
gpus
:- Type: INT
- Default:
0
- Usage:
--gpus
Number of GPUs needed for the service.
-
gpu_type
:- Type: Choice(’[‘V100’, ‘A10G’, ‘T4’, ‘L4’]‘)
- Default:
none
- Usage:
--gpu-type
Type of GPU.
-
cpu
:- Type: FLOAT
- Default:
100
- Usage:
--cpu
Number of CPU millicores. 1000 = 1 CPU
-
memory
:- Type: FLOAT
- Default:
200
- Usage:
--memory
Amount of RAM in megabytes.
-
min_scale
:- Type: INT
- Default:
0
- Usage:
--min-scale
Minimum number of pods to run.
-
max_scale
:- Type: INT
- Default:
3
- Usage:
--max-scale
Maximum number of pods to run.
-
env
:- Type: STRING
- Default:
default
- Usage:
--env
Environment to deploy in.
-
secret
- Type: STRING
- Multiple: True
- Default:
- Usage:
--secret
-
config_file
:- Type: TEXT
- Default:
none
- Usage:
--config-file
Path to the config.yaml file. You can use this file instead of cli flags. Keys in config.yaml take precedence than cli flags..
-
help
:- Type: BOOL
- Default:
false
- Usage:
--help
Show this message and exit.