Advanced Topics

In this section we cover some more advanced topics around installation and configuration.

Config file

The config file is created the first time you run python -m remo_app init.

It gets stored in: HOME_DIRECTORY/.remo/remo.json

Location of your HOME directory depends upon your specific OS system.

Content:

{
  "conda_env": "path-to-conda-environment-where-you-installed-remo-app",
  "port": "8000",
  "server": "http://localhost",
  "user_email": "[email protected]",
  "user_name": "your-username",
  "user_password": "adminpass",
  "viewer": "electron"
}

NB: Changing some values would require running again python -m remo_app update in order for the changes to be processed.

Viewer

Accepted values:

  • electron
  • browser
  • jupyter

By default, Remo uses Electron to run the app. This allows for more vertical space compared to using Remo in a browser.

If you prefer to access Remo from a browser (Remo was tested with Chrome), you can directly browse to where Remo is being served (default: localhost:8123). You can also change the value of the viewer in the config file to browser, and on launch a browser window will pop open.


Uploading Data

When possible, Remo tries to link your local data in order to minimize needed extra storage. However, adding data still takes a bit of extra storage space. See below for more details.

Linking datasets

Images are linked and opened directly from your hard disk if you upload data using the Electron app and if they are not archive or annotation files.

This however means that if you move or delete your images after uploading them in Remo, Remo will fail to find them.

NB: even when linking data, Remo still creates a copy of smaller version of the images to show thumbnails and previews in an efficient way. Those shouldn't amount to more than 15% of the original size of the dataset.

Copying data

Annotations are automatically uploaded as new data on Remo. Image data will also be uploaded as a copy if it's passed as an archive, or if using the file picker directly from the Browser.


Database

We currently support only PostsgreSQL.

Support for MySQL is in the plans at some point.


Installation issues

In this section we will collect some troubleshooting steps that can help with installation issues.

In case you are stuck, feel free to check existing issues on our forum or open a new discussion.

Ubuntu

On a fresh Ubuntu machine, depending on what's installed you may need to install gcc and python3-dev packages.

PostgreSQL Installation

When doing python -m remo_app init, the script will try to automatically install PostgreSQL.

If the installation fails, you can try to install PostgreSQL manually (e.g. as described here: https://www.postgresqltutorial.com/install-postgresql/ ) and then run again python -m remo_app init.


Remo version

You can check your version of remo by either:

  • pointing the mouse to the Remo icon within the webapp
  • calling python -m remo_app --version

medium_width_pic


Docker Support WIP

We are planning to release Docker images for remo at some point. If this is something you are particularly interested in, let us know and we can try to prioritise it.