exogibbs.io package

Submodules

exogibbs.io.load_data module

exogibbs.io.load_data.get_data_filepath(filename)

get the full path of the data file

Parameters:

  • filename (str) – filename of the test data

  • dirname (str) – directory name of the test data (default: “data/testdata/”)

Returns:

full path of the test data file

Return type:

str

exogibbs.io.load_data.load_JANAF_molecules(df_molname: DataFrame, path_JANAF_data: str | Path, *, tag: str = '(g)', save_hdf5: str | None = None, hdf5_compression: str = 'zlib') → Dict[str, DataFrame]

Load JANAF tables for all molecules listed in df_molname and return them as a dict {molecule: DataFrame}. Each DataFrame keeps its original (row, col) shape, which is convenient for later conversion to JAX arrays.

Parameters:

  • df_molname – pd.DataFrame Must contain a column named "Molecule" with file prefixes.

  • path_JANAF_data – str or Path Directory that holds <molecule>(g).txt files.

  • tag – str tag for JANAF file default to “(g)”

  • save_hdf5 – str, optional If given, the dict is additionally written to an HDF5 file whose key is the molecule name (e.g. /H2O). Passing None skips on-disk storage.

  • hdf5_compression – {“zlib”, “blosc”, …}, default “zlib” Compression algorithm for the HDF5 table format.

Returns:

dict[str, pd.DataFrame] Mapping from molecule name to its JANAF table.

Notes

  • Missing files or read errors are skipped with a warning.

  • Use jax.numpy.asarray(df.to_numpy()) when you need a JAX array.

Examples

>>> df_molname = load_molname_ykb4()
>>> path_JANAF_data = "/home/kawahara/thermochemical_equilibrium/Equilibrium/JANAF"
>>> matrices = load_JANAF_molecules(df_molname, path_JANAF_data)
>>> mat = matrices["C1O2"].to_numpy()
exogibbs.io.load_data.load_JANAF_rawtxt(filename)

loads a JANAF raw text file

Parameters:

filename (str) – filename of the JANAF raw text file, e.g. ‘H2(g).txt’

Returns:

DataFrame containing the data from the JANAF raw text file

Return type:

pd.DataFrame

Module contents