msl.io.dataset module

A Dataset is essentially a numpy.ndarray with Metadata.

class msl.io.dataset.Dataset(name, parent, read_only, shape=(0, ), dtype=<class 'float'>, buffer=None, offset=0, strides=None, order=None, data=None, **metadata)[source]

Bases: Vertex

A Dataset is essentially a numpy.ndarray with Metadata.

Do not instantiate directly. Create a new Dataset using create_dataset().

Parameters:

name (str) – A name to associate with this Dataset.
parent (Group) – The parent Group to the Dataset.
read_only (bool) – Whether the Dataset is to be accessed in read-only mode.
shape – See numpy.ndarray
dtype – See numpy.ndarray
buffer – See numpy.ndarray
offset – See numpy.ndarray
strides – See numpy.ndarray
order – See numpy.ndarray
data – If not None then it must be either a numpy.ndarray or an array-like object which will be passed to numpy.asarray(), as well as dtype and order, to be used as the underlying data.
**metadata – All other key-value pairs will be used as Metadata for this Dataset.

copy(read_only=None)[source]

Create a copy of this Dataset.

Parameters:: read_only (bool, optional) – Whether the copy should be created in read-only mode. If None then creates a copy using the mode for the Dataset that is being copied.
Returns:: Dataset – A copy of this Dataset.

property data

The data of the Dataset.

Note

You do not have to call this attribute to get access to the numpy.ndarray. You can directly call the numpy.ndarray attribute from the Dataset object.

For example,

>>> dset
<Dataset '/my_data' shape=(4, 3) dtype='<f8' (0 metadata)>
>>> dset.data
array([[ 0.,  1.,  2.],
       [ 3.,  4.,  5.],
       [ 6.,  7.,  8.],
       [ 9., 10., 11.]])
>>> dset.size
12
>>> dset.tolist()
[[0.0, 1.0, 2.0], [3.0, 4.0, 5.0], [6.0, 7.0, 8.0], [9.0, 10.0, 11.0]]
>>> dset.mean(axis=0)
array([4.5, 5.5, 6.5])
>>> dset[::2]
array([[0., 1., 2.],
       [6., 7., 8.]])

Type:: numpy.ndarray

property read_only

Whether this Dataset is in read-only mode.

This is equivalent to setting the WRITEABLE property in numpy.ndarray.setflags().

Type:: bool