filters#

catlas.filters.adsorbate_filter(config, dask_df, sankey)#

Filters adsorbate structures according to rules specified in a config.

Parameters

  • config (dict) – dictionary specifying what criteria to filter on

  • dask_df (dask.dataframe.core.DataFrame) – adsorbates to filter

  • sankey (catlas.sankey.sankey_utils.Sankey) – a sankey object to update as adsorbates are filtered

Returns

a dataframe of filtered adsorbates catlas.sankey.sankey_utils.Sankey: the input sankey object with added information

Return type

dask.dataframe.core.DataFrame

catlas.filters.bulk_filter(config, dask_df, sankey=None, initial_bulks=None)#

Filters a dask dataframe of bulk structures according to rules specified in a config yml.

Parameters

  • config (dict) – dict describing what bulks to filter

  • dask_df (dask.core.frame.DataFrame) – bulk materials to select from

  • sankey (catlas.sankey.sankey_utils.Sankey) – object that records the number of bulks and filtered bulks

  • initial_bulks (int) – the initial number of bulks

Returns

filtered bulk materials catlas.sankey.sankey_utils.Sankey: the input Sankey object with added info

Return type

dask.core.frame.DataFrame

catlas.filters.predictions_filter(bag_partition, config)#

Filter surfaces based on their predicted adsorption energies.

Parameters

  • bag_partition (Iterable[dict]) – a partition of a Dask Bag. Each partition should contain columns “bulk_id”, “slab_millers”, “slab_shift”, “slab_top”, “adsorbate_smiles”, “filter_reason”, and a column whose name is the same as the value in the “filter_column” field of the input config.

  • config (dict) – a dictionary specifying how to filter predictions

  • sankey (catlas.sankey.sankey_utils.Sankey) – a Sankey object to update with prediction filters.

Returns

an updated version of the input Bag partition. Contains an updated

column “filter_reason” that has been modified to include rows filtered out according to the input config.

Return type

Iterable[dict]

catlas.filters.slab_filter(bag_partition, config)#

Filters a slab according to rules specified in a config.

Parameters

  • bag_partition (Iterable[dict]) – unfiltered bag partition of enumerated slabs

  • config (dict) – dict containing slab filtering criteria

Returns

filtered bag partition

Return type

Iterable[dict]