- Generated by
1.14.0
|
sgsPy
structurally guided sampling
|
Functions | |
| sgspy.calculate.distribution.distribution.distribution (SpatialRaster rast, Optional[str|int] band=None, Optional[SpatialVector] samples=None, Optional[str] layer=None, int bins=50, bool plot=True) | |
| This function calculates the distribution of a given raster band. | |
| sgspy.calculate.distribution.distribution.distribution | ( | SpatialRaster | rast, |
| Optional[str | int] | band = None, | ||
| Optional[SpatialVector] | samples = None, | ||
| Optional[str] | layer = None, | ||
| int | bins = 50, | ||
| bool | plot = True ) |
This function calculates the distribution of a given raster band.
The distribution is plotted by default, it is also returned.
A number of bins can be given (the default is 50), the bins will be sized equally depending on the minimum and maximum values within the raster band. If the raster is a multi-band raster, the 'band' parameter must be passed.
Additionally, a SpatialVector may be passed to the samples parameter which must contain a sample plot network (only geometry types Point or MultiPoint). If a SpatialVector is passed this way, the distribution of the samples will also be calculated, using the same bins as the layer distribution. if the SpatialVector has more than one layer, the 'layer' parameter must be passed.
This function returns a dict, the keys of the dict will be either 'population' or 'sample'. If the 'samples' parameter is not given, only the 'population' key will exist. If the 'samples' parameter IS given, both 'population' and 'sample' will exist as keys. The values of the dict will be two arrays, one is the array of bins of type float64, the other is the array of counts of type int64. The array of bins will be one value longer than the array of counts, as it countains not just the minimum value but also the maximum value.
For example, if the minimum pixel value in the raster is 1, and the maximum pixel value in the raster is 3, 3 and the bin size is 4, then the bin array would be [1.0, 1.5, 2.0, 2.5, 3.0]. The array of counts would then be of length 4, where the first element (index 0) would be the count of pixels which have a pixel value that satisfies the following 1.0 <= val < 1.5. The second element (index 1) would then be 1.5 <= val < 2.0, and so on until the final element with 2.5 <= val <= 3. This is typical, and matches the way that the numpy histogram function works.
rast = sgspy.SpatialRaster("rast.tif") #single band raster
sgspy.calculate.distribution(rast) #plotting is default
rast = sgspy.SpatialRaster("mraster.tif")
sgspy.calculate.distribution(rast, band='pzabove2', bins=10)
rast = sgspy.SpatialRaster("mraster.tif")
samples = sgspy.sample.clhs(rast, num_samples=250)
sgspy.calculate.distribution(rast, band='zsd', samples=samples)
rast = sgspy.SpatialRaster("mraster.tif")
samples = sgpsy.sample.clhs(rast, num_samples=240)
result = sgspy.calculate.distribution(rast, band='zq90', samples=samples, bins=100, plot=False)
[pop_bins, pop_counts] = result['population']
[samp_bins, samp_counts] = result['sample']
rast : SpatialRaster
raster data structure containing input raster band
band : Optional[str | int]
the band to use within 'rast' if 'rast' has more than one band
samples : Optional[SpatialVector] the option data structure containing an input sample network
layer : Optional[str]
the layer name of the sample network if 'samples' contains more than one layer
bins : int
the number of bins in the histogram distribution
plot : bool
whether to plot a histogram of the distribution
a dict[str, (array, array)] containing the bin values and counts, with dict keys 'population' and potentially 'sample'
1.14.0