AESOpt Param (aesoptparam)

Is a light weight extension of the open source param project.

The param project is a library for handling all the user-modifiable parameters, arguments, and attributes that control code. It adds a way to declare parameters via their type while also adding meta data such as documentation, valid range, etc. For a more extensive introduction user are referred to the project homepage: https://param.holoviz.org/ The param webpage is also where users are expected to look for more detailed use of aesoptparam.

aesoptparam is purely extending param, by adding the following features:

• Make an extended Parameterized -> AESOptParameterized, which adds the following features:
  • Method for getting data as a dict (.as_dict)
  • Method for serializing data (.as_serial). Used for saving data to JSON, YAML, TOML
  • Method for setting data from a dict (.from_dict). Also works for nested dicts
  • Method for validating numpy array shapes (.validate_array_shapes). See AESOptArray for more.
  • Method for interpolating arrays (.interp)
  • Method for setting values in others units than their default units (.set_val)
  • Method for gettting values in others units than their default units (.get_val)
  • HTML rendering of the AESOptParameterized instance content (Name, Documentation, Type, Range, Units, Value)
  • Adds posibility for setting a parent_object for nested AESOptParameterized, which allows others values to refer to any other value however deeply nested it is
• A set of new/modified parameters:
  • SubParameterized: Parameter to add nested AESOptParameterized. It will automatically add the parent object.
  • ListOfParameterized: Parameter to add a list of AESOptParameterized.
  • AESOptString: Adds the following list of attributes or methods:
    • default_ref: Referencing another parameter string
  • AESOptBoolean: Adds the following list of attributes or methods:
    • default_ref: Referencing another parameter string
  • AESOptNumber: Adds the following list of attributes or methods:
    • units
    • default_ref: Referencing another parameter number
  • AESOptInteger: Adds the following list of attributes or methods:
    • units
    • default_ref: Referencing another parameter number
  • AESOptArray: Adds the following list of attributes or methods:
    • bounds, softbounds. inclusive_bounds, similar to param.Number
    • units
    • dtype: The array data type
    • shape: The shape of the array. Can also be a reference to another parameter. Used for validating array shape.
    • default_full: Makes the default a array full of a given number. Using numpy.full. The first argument can reference others parameters to get their shape.
    • default_interp: The default is to interpolate the array from other arrays. Arguments can both be arrays or references to others arrays.
    • default_ref: Referencing another parameter array

Besides the core features above, it also adds some utility tools:

• display_json_data : Function to render JSON compliant data as a HTML table in Jupyter notebooks. It also works for numpy data that has numeric data dtypes.
• read_json : Method for reading JSON data file, which automatically can convert numeric list's to numpy.
• write_json : Method for writing JSON data file, which automatically can convert numeric numpy arrays and types to list of build-in types.

Documentation

AESOpt Param Documentation