Skip to content

YAML File Validation ​

The entire pem workflow requires a significant number of parameters and constants to run. These are all gathered in a YAML file, which is validated through a pydantic process. When setting up a YAML file for a project for the first time, we recommend using the interface. Once saved, you can edit the YAML file in any standard text editor. The interface hides certain standard settings that typically do not require modification. Example files with all options and commented standard settings can be found in the GitHub repository.

For porosity adjustment, the interface selects the method field through the dropdown and includes it in the exported YAML (none, constant, or grid_file). This field is hidden in the form but remains in the JSON Schema so validation can distinguish the three options.

The interface accepts underscores as decimal digit separators in numeric YAML values, such as 25_000_000_000.0, matching the Python configuration loader. Quoted values remain strings. Exported numbers may omit the separators.

There is also a global configuration file in YAML format, which is used throughout the FMU process. In this context, the main information used is a set of dates for which the estimation is made, also the dates from which difference seismic is estimated. The name of output grid is also read from the global config file.

Validation Process ​

Entries in the YAML file are validated through pydantic. This validation checks that referenced files and directories exist, that parameters are of the correct type, and to some extent that numerical values fall within expected ranges. If pydantic encounters errors, messages will appear in the console or in error message files if the process is run from ert. Unfortunately, the pydantic error messages can be cryptic, and you may need to contact an fmu-pem developer or superuser to resolve the issue.