csdmpy

The csdmpy is a python package for importing and exporting files serialized with the core scientific dataset model file-format. The package supports a \(p\)-component dependent variable, \(\mathbf{U} \equiv \{\mathbf{U}_{0}, \ldots,\mathbf{U}_{q}, \ldots,\mathbf{U}_{p-1} \}\), which is discretely sampled at \(M\) unique points in a \(d\)-dimensional space \((\mathbf{X}_0, \ldots \mathbf{X}_k, \ldots \mathbf{X}_{d-1})\). Besides, the package also supports multiple dependent variables, \(\mathbf{U}_i\), sharing the same \(d\)-dimensional space.

Here, every dataset is an instance of the CSDM class, which holds a list of dimensions and dependent variables. Every dimension, \(\mathbf{X}_k\), is an instance of the Dimension class, while every dependent variable, \(\mathbf{U}_i\), is an instance of the DependentVariable class.

Methods

Methods Summary

load([filename, application])	Load a .csdf/.csdfe file and return an instance of CSDM class.
new([description])	Create a new instance of the CSDM class containing a 0D{0} dataset.
parse_dict(dictionary)	Parse a CSDM compliant python dictionary and return a CSDM object.
loads(string)	Loads a JSON serialized string as a CSDM object.
plot(data_object, **kwargs)	Helper function for plotting basic 1D and 2D datasets.

Method Documentation

csdmpy.load(filename=None, application=False)

Load a .csdf/.csdfe file and return an instance of CSDM class.

The file must be a JSON serialization of the CSD Model.

Example

>>> data1 = cp.load('local_address/file.csdf') 
>>> data2 = cp.load('url_address/file.csdf') 

Parameters

• filename (str) – A local or remote address to the .csdf or `.csdfe file.

• application (bool) – If true, the application metadata from application that last serialized the file will be imported. Default is False.

• sort_fft_order (bool) – If true, the coordinates and the components corresponding to the dimension with complex_fft as True will be sorted upon import and the corresponding complex_fft key-value will be set to False. Default is True.

Returns

A CSDM instance.

csdmpy.new(description='')

Create a new instance of the CSDM class containing a 0D{0} dataset.

Parameters

description (str) – A string describing the the csdm object. This is optional.

Example

>>> import csdmpy as cp
>>> emptydata = cp.new(description='Testing Testing 1 2 3')
>>> print(emptydata.data_structure)
{
  "csdm": {
    "version": "1.0",
    "description": "Testing Testing 1 2 3",
    "dimensions": [],
    "dependent_variables": []
  }
}

Returns

A CSDM instance.

csdmpy.parse_dict(dictionary)

Parse a CSDM compliant python dictionary and return a CSDM object.

Parameters

dictionary – A CSDM compliant python dictionary.

csdmpy.loads(string)

Loads a JSON serialized string as a CSDM object.

Parameters

string – A JSON serialized CSDM string.

Returns

A CSDM object

Example

>>> object_from_string = cp.loads(cp.new('A test dump').dumps())
>>> print(object_from_string.data_structure)  
{
  "csdm": {
    "version": "1.0",
    "timestamp": "2019-10-21T20:33:17Z",
    "description": "A test dump",
    "dimensions": [],
    "dependent_variables": []
  }
}

csdmpy.plot(data_object, **kwargs)

Helper function for plotting basic 1D and 2D datasets.

Parameters

• data_object – The csdm object

• reverse_axis – An array of boolean specifying which dimensions will be displayed in reverse.

• kwargs – Additional kwargs which are passed to the matplotlib plot or imshow functions.

Example

>>> cp.plot(data_object)