Settings and the Download Location

Configuration for em_database lives in a live, matplotlib-rcParams-style object, em_database.settings, seeded at import from a YAML file in ~/.em_database/settings.yaml. This example shows how to read it, change the data directory for the session or for good, and reset it.

import em_database

The data directory is where datasets download to. It defaults to ~/em_database.

print("Current download directory:", em_database.get_data_dir())

Current download directory: /home/runner/em_database

Change it for this session only (in memory, not written to disk).

em_database.set_data_dir("/path/to/scratch", persist=False)
print("Session directory:", em_database.get_data_dir())

Session directory: /path/to/scratch

Change it and remember the choice across sessions (writes ~/.em_database/settings.yaml). set_data_dir persists by default; the equivalent low-level form is em_database.settings[...] = ...; save().

em_database.set_data_dir("/big/disk/em_data")            # set + persist
# em_database.settings["data_dir"] = "/big/disk/em_data"  # the same thing
# em_database.settings.save()
print("Persisted directory:", em_database.get_data_dir())

Persisted directory: /big/disk/em_data

Reset to the default location, forgetting the saved choice.

em_database.reset_data_dir()
print("Reset directory:", em_database.get_data_dir())

Reset directory: /home/runner/em_database

Any setting can be stored, not just the data directory.

em_database.set_setting("quality", "high")
print("quality:", em_database.get_setting("quality"))

quality: high

Datasets can also be installed system-wide and shared by every user. Point EM_DATABASE_SHARED_DIR (or a shared_data_dirs setting, or a system config file) at the shared location, and download() / filepath() will find a pre-installed copy there before downloading into your own directory.

Clean up so running this example leaves no settings behind.

em_database.settings.reset()