JADE-RAW-RESULTS

This repo hosts the raw data coming from the post-processing of JADE assessments.

Uploading new results

For the vast majority of cases, adding data to this repository is pretty straightforward, the .csv files that are output by JADE in the Raw_Data folders, can be directly copied in the corresponding folder in this repository structure, which looks like the following:

ROOT
  |-----<lib1>
  |       |-----<>
  |
  |-----<lib2>
          |-----<benchmark_name>
                      |-----------<mcnp>
                      |             |------<Raw_Data>
                      |                        |--------<raw_data1>.csv
                      |                        |--------<raw_data2>.csv
                      |                        |--------<...>
                      |                        |--------metadata.json
                      |
                      |-----------<openmc>
                      |-----------<...>

Some sanity checks are performed on the metadata.json keywords and values to ensure they are consistent with the allowed library names and with the folder structure they are inserted into (for more details check the CI tests).

If a large set of data needs to be added, copying by hand each folder can become a tedious operation. import_all.py can be used to automatically scan a JADE > v3.0.0 post-processing folder, identify which set of raw data are not currently available in the database and add them to it.

Note

If a new library is added, remember to update the import_all.py and metadata_test.py files to provide the default name for the new library.

Exceptions

Tiara-BS

The .csv raw data structure in this benchmark is not compatible with JADE results. To correctly import them into this repository use this import_tiara_bs.py.

Tiara-FC

The .csv raw data structure in this benchmark is not compatible with JADE results. To correctly import them into this repository use this import_tiara_fc.py.

metadata.json

The metadata.json file contains important information about the results set. It's structured as a JSON object as seen for the example below.

{
    "jade_version": "3.0.0",
    "code": "mcnp",
    "code_version": "6.2",
    "library": "FENDL 3.2b",
    "benchmark_name": "ASPIS-Fe88",
    "benchmark_version": "1.0",
    "jade_run_version": "3.0.1",
    # optional, this makes sense only for d1s assessments
    "transport_lib": "FENDL 3.1d",
}

• jade_version: The version of JADE used to perform the post processing.
• code: The transport code used to run the simulations (in this case, MCNP).
• code_version: The version number of the transport code.
• library: The name of the nuclear data library used. (activation lib in case of a d1s asessment)
• benchmark_name: The name of the experimental or computational benchmark.
• benchmark_version: The version of the benchmark in JADE.
• jade_run_version: The version of JADE that was used to run the benchmark.
• transport_lib: Optional, the transport lib used during a D1S assessment.