Input Arrays

Many pfdf commands require one or more numeric arguments as input. The library ultimately converts these inputs to numpy arrays, but pfdf is flexible enough to accept a variety of input types. Here, we distinguish between scalar, vector, and matrix inputs.

Scalars

A scalar input has a single data element. The following values are all acceptable scalars:

import numpy as np

# Numpy scalars, or arrays with a single element
np.array(5)
np.array(5).reshape(1)
np.array(5).reshape(1,1,1,1,1,1,1)

# int or float
5
5.0

# List or tuple with a single element
[5]
(5,)

Vectors

A vector is a 1D array of data values. In the context of pfdf, a vector input may have any number of dimensions, but only one dimension may be non-singleton (have a length greater than 1). The following values are all acceptable vectors:

import numpy as np

# All scalars
np.array(5)
5
[5]
# etc.

# Numpy arrays with one non-singleton dimension
np.arange(5)
np.arange(5).reshape(1,1,1,1,5,1,1)

# List or tuple of ints/floats
[1,2,3.3,4,5.5]
(1,2,3.3,4,5.5)

Matrices

A matrix is a 2D array of data values. In the context of pfdf, most matrices are handled via the Raster class. However, when this is not the case, a matrix should be an array in which only the first two dimensions may be non-singleton. The following values are acceptable matrices:

import numpy as np

# All scalars and most vectors
5
[5, 6, 7]
np.arange(5)
# etc.

# Only dimensions<2 are non-singleton
np.arange(10).reshape(10,1)
np.arange(10).reshape(5,2)
np.arange(10).resshape(5,2,1,1,1,1,1)