To deploy the system locally using docker-compose, follow those steps:
Access: https://github.com/VIDA-NYU/ptg-api-server
To bring up the entire server on your local machine, do:
make full
To bring up everything except for the machine learning containers (in case you're using a different machine for GPUs), you can do:
make api
make dash
make record
then on the other machine, you would run:
make ml
For more information about how to set up the server environment click here.
Ports:
- The ARGUS dashboard is at http://localhost:3000
- The api is at http://localhost:7890 - for information about how to interact with the API, see:
- CLI and Python client: https://github.com/VIDA-NYU/ptgctl
- API Docs: https://api.ptg.poly.edu/docs
- To interact with the mongodb instance, you can use Mongo Express - a GUI for editing documents: - Mongo express is available at http://localhost:8081
To setup the Hololens, you need to configure the URL for the API server. Currently to do this, it involves going into the Hololens source code and changing the URL that the Hololens is pointing to. Obviously this is not great.
Perhaps there is a way to have a textbox/dropdown for reconfiguring the URL in the Hololens UI?
make down