PhysicalPropertyDataSet
- class openff.evaluator.datasets.PhysicalPropertyDataSet[source]
An object for storing and curating data sets of both physical property measurements and estimated. This class defines a number of convenience functions for filtering out unwanted properties, and for generating general statistics (such as the number of properties per substance) about the set.
Methods
__init__
()Constructs a new PhysicalPropertyDataSet object.
add_properties
(*physical_properties[, validate])Adds a physical property to the data set.
from_json
(file_path)Create this object from a JSON file.
from_pandas
(data_frame)Constructs a data set object from a pandas
DataFrame
object.json
([file_path, format])Creates a JSON representation of this class.
merge
(data_set[, validate])Merge another data set into the current one.
parse_json
(string_contents)Parses a typed json string into the corresponding class structure.
properties_by_substance
(substance)A generator which may be used to loop over all of the properties which were measured for a particular substance.
properties_by_type
(property_type)A generator which may be used to loop over all of properties of a particular type, e.g.
Converts a PhysicalPropertyDataSet to a pandas.DataFrame object with columns of
validate
()Checks to ensure that all properties within the set are valid physical property object.
Attributes
A list of all of the properties within this set.
The types of property within this data set.
The sources from which the properties in this data set were gathered.
The substances for which the properties in this data set were collected for.
- property properties
A list of all of the properties within this set.
- Type
tuple of PhysicalProperty
- property property_types
The types of property within this data set.
- Type
set of str
- property substances
The substances for which the properties in this data set were collected for.
- Type
set of Substance
- property sources
The sources from which the properties in this data set were gathered.
- Type
set of Source
- merge(data_set, validate=True)[source]
Merge another data set into the current one.
- Parameters
data_set (PhysicalPropertyDataSet) – The secondary data set to merge into this one.
validate (bool) – Whether to validate the other data set before merging.
- add_properties(*physical_properties, validate=True)[source]
Adds a physical property to the data set.
- Parameters
physical_properties (PhysicalProperty) – The physical property to add.
validate (bool) – Whether to validate the properties before adding them to the set.
- properties_by_substance(substance)[source]
A generator which may be used to loop over all of the properties which were measured for a particular substance.
- Parameters
substance (Substance) – The substance of interest.
- Returns
- Return type
generator of PhysicalProperty
- properties_by_type(property_type)[source]
A generator which may be used to loop over all of properties of a particular type, e.g. all “Density” properties.
- Parameters
property_type (str or type of PhysicalProperty) – The type of property of interest. This may either be the string class name of the property or the class type.
- Returns
- Return type
generator of PhysicalProperty
- validate()[source]
Checks to ensure that all properties within the set are valid physical property object.
- to_pandas()[source]
Converts a PhysicalPropertyDataSet to a pandas.DataFrame object with columns of
‘Id’
‘Temperature (K)’
‘Pressure (kPa)’
‘Phase’
‘N Components’
‘Component 1’
‘Role 1’
‘Mole Fraction 1’
‘Exact Amount 1’
…
‘Component N’
‘Role N’
‘Mole Fraction N’
‘Exact Amount N’
‘<Property 1> Value (<default unit>)’
‘<Property 1> Uncertainty / (<default unit>)’
…
‘<Property N> Value / (<default unit>)’
‘<Property N> Uncertainty / (<default unit>)’
‘Source’
where ‘Component X’ is a column containing the smiles representation of component X.
- Returns
The create data frame.
- Return type
- classmethod from_json(file_path)
Create this object from a JSON file.
- Parameters
file_path (str) – The path to load the JSON from.
- Returns
The parsed class.
- Return type
cls
- classmethod from_pandas(data_frame: pandas.core.frame.DataFrame) openff.evaluator.datasets.datasets.PhysicalPropertyDataSet [source]
Constructs a data set object from a pandas
DataFrame
object.Notes
All physical properties are assumed to be source from experimental measurements.
Currently this method onlu supports data frames containing properties which are built-in to the framework (e.g. Density).
This method assumes the data frame has a structure identical to that produced by the
PhysicalPropertyDataSet.to_pandas
function.
- Parameters
data_frame – The data frame to construct the data set from.
- Returns
- Return type
The constructed data set.
- json(file_path=None, format=False)
Creates a JSON representation of this class.