Command line reference

Installing mozo installs the mozo command. mozo --help lists the commands and mozo <command> --help the options of one.

mozo start

Bash
mozo start [--host HOST] [--port PORT] [--workers N] [--reload]

Start the HTTP server.

Option Default Description
--host 0.0.0.0 Address to listen on.
--port 8000 Port to listen on.
--workers 1 Worker processes. Each loads its own copy of every model it serves, so memory grows with each one.
--reload off Restart when the code changes, for development. Forces one worker.

The server reads MOZO_ENABLE, MOZO_CACHE, MOZO_OFFLINE, MOZO_BASE_URL and MOZO_PREPARED_MB from the environment; Configuration has them. HTTP server is the guide.

mozo run

Bash
mozo run WORKFLOW [--file PATH] [--set NAME=VALUE ...] [--test]

Run a workflow from a JSON file, with no browser and no server. It runs over the whole source: every image in a folder, every frame of a video. Progress is printed as it goes; nodes that write files have written them when it ends.

Option Description
--file Run on this image, video or folder instead of the one saved in the workflow.
--set NAME=VALUE Override a parameter, by name or as node.name for one node's. Repeatable.
--test Run one item and print what every node produced, instead of the whole source.
Bash
mozo run blur_faces.json --file example.jpg --set threshold=0.5
Text only
Blur faces -- Find every face and blur it.
  1 item in 2.8s (0/s)

--test describes each node's output in a line: an image's size, how many detections and of what, a depth map's range. Use it to check a graph does its job before you point it at ten thousand files:

Text only
Blur faces -- Find every face and blur it.
  read: a 1920x1281 image
  find: 5 x ['face']
  blur: a 1920x1281 image
  save: nothing (an end)

Run a workflow covers the same from Python.

mozo version

Bash
mozo version
Text only
Mozo version 1.10.0