Run Auto-Editor on the Golem Network.
To run you first need to have yagna installed and the yagna daemon running. Instructions on how to do that here and here.
Running the program looks like:
python3 requstor.py INPUT_FILE [--auto-editor-args="[ARGS]"] [--output ]
INPUT_FILE is the file you want processed. Although auto-editor supports multiple files as input, this program currently does not. Additionally, any args regarding output to in the auto-editor-args field will be ignored, as those will be overwritten by the program.
Be careful when specifying video codecs to auto-editor, as only those that ship by default with ffmpeg on debian will be present. Most common codecs like h26x and vpx are supported. Full list can be found here.
for more info run python3 requestor.py --help
so far I have only tested it on ubuntu 20.04
- python packages:
- yapapi
- ffmpeg and ffprobe
if you want to modify or build the docker image yourself you will need some additional dependancies:
- docker
- gvmkit-build
Auto-Editor is a command line tool to automate some simple video editiing tasks, such as cutting out silence and dead-space. Great for students who want to watch lecture videos in less time, or content creators editing a long interview.
you can absolutely run auto-editor on your own computer, but the video encoding and other processing that it performs can be very intensive, so distributing it over many computers can help speed it up greatly. Or, it can make processing of long videos feasible on lower spec machines, potentially even a Chromebook. See benchmark.md for an example of the speedup.
Golem Network is a decentralized distributed super-computing platform. Read more about it at their website: https://golem.network.