schrodinger.application.models.execution.sources module

class schrodinger.application.models.execution.sources.AbstractRowSource

Bases: object

generateRows() → Iterator[dict[str, Any]]

Must be implemented in child classes as a generator that yields dicts representing rows, where each dict contains the row data as well as a unique identifier under the key ROW_SOURCE_ID_KEY.

getRowSchema() → dict[str, type]

Returns a dictionary mapping input names to their types for the rows generated by this row source. This is used to infer the schema of the rows.

class schrodinger.application.models.execution.sources.ListRowSource(rows)

Bases: AbstractRowSource

A row source that reads row dicts from a list, which is supplied at initialization. The list index is used as the unique row identifier.

__init__(rows)

generateRows()

Must be implemented in child classes as a generator that yields dicts representing rows, where each dict contains the row data as well as a unique identifier under the key ROW_SOURCE_ID_KEY.

getRowSchema() → dict[str, type]

Returns the types of the row dicts in the list.

class schrodinger.application.models.execution.sources.StructureFileRowSource(struct_file, property_map)

Bases: AbstractRowSource

A row source that reads structures from a structure file and yields a row dict for each structure. The index of the structure in the file is used as the unique row identifier.

In addition to the structure filename, a mapping of model input names to structure properties must be specified. A special value STRUCTURE can be used as a property to indicate that the structure itself should be included in the row dict under the corresponding input name.

Example mapping:

{‘foo’: ‘s_m_foo’, ‘bar’: ‘i_m_bar’, ‘ligand’: STRUCTURE }

STRUCTURE = <object object>

__init__(struct_file, property_map)

generateRows()

Must be implemented in child classes as a generator that yields dicts representing rows, where each dict contains the row data as well as a unique identifier under the key ROW_SOURCE_ID_KEY.

getRowSchema() → dict[str, type]

Returns a dictionary mapping input names to their types for the rows generated by this row source.

class schrodinger.application.models.execution.sources.LegacyLDRowSource(input_file_tuples: list[tuple[ColumnInput, str]])

Bases: AbstractRowSource

A row source that reads column input data from CSV files as created by the legacy LiveDesign models. Each CSV file corresponds to a column input, and the rows are identified by a corporate ID. The row source yields a row dict with the corporate ID as the unique identifier. The csv file for each input must be passed in as a tuple of (ColumnInput, filename) pairs.

Example input file tuples:

(CompositeInput(name=’smiles’, type=InputType.STRING), ‘file1.csv’) (CompositeInput(name=’charge’, type=InputType.INTEGER), ‘file2.csv’)

SDF_FILE = 'SDF_FILE'

__init__(input_file_tuples: list[tuple[ColumnInput, str]])

generateRows()

Must be implemented in child classes as a generator that yields dicts representing rows, where each dict contains the row data as well as a unique identifier under the key ROW_SOURCE_ID_KEY.

getRowSchema() → dict[str, type]

Returns a dictionary mapping input names to their types for the rows generated by this row source.