This section focuses on the Morpheus CLI and illustrates how the CLI can be used to configure and run a Morpheus Pipeline.
The Morpheus CLI is built on the Click Python package which allows for nested commands and chaining multiple commands together. At a high level, the CLI is broken up into two main sections:
run
- For running NLP or FIL pipelines.
tools
- Tools/Utilities to help set up, configure and run pipelines and external resources
Users can get help on any command by passing --help
to a command or sub-command. For example, to get help on the
tools:
$ morpheus tools --help
Usage: morpheus tools [OPTIONS] COMMAND [ARGS]...
Options:
--help Show this message and exit. [default: False]
Commands:
autocomplete Utility for installing/updating/removing shell completion for
Morpheus
onnx-to-trt Converts an ONNX model to a TRT engine
The help text will show arguments, options and all possible sub-commands. Help for each of these sub-commands can be queried in the same manner:
$ morpheus tools onnx-to-trt --help
Usage: morpheus tools onnx-to-trt [OPTIONS]
Options:
--input_model PATH [required]
--output_model PATH [required]
--batches <INTEGER INTEGER>... [required]
--seq_length INTEGER [required]
--max_workspace_size INTEGER [default: 16000]
--help Show this message and exit. [default:
False]
The Morpheus CLI supports bash, fish, zsh, and powershell autocompletion. To set up autocomplete, it must first be installed. Morpheus comes with a tool to assist with this:
$ morpheus tools autocomplete install
bash completion installed in ~/.bash_completion
$ source ~/.bash_completion
After autocomplete has been installed, [TAB]
can be used to show all commands, options and arguments when building
pipelines via the CLI:
$ morpheus run pipeline- # [TAB][TAB]
pipeline-fil pipeline-nlp