Merge pull request #24 from TerrainBento/barnhark/create_mask

Barnhark/create mask

docs/source/umami.calculations.metric.mask_aggregation.rst

@@ -0,0 +1,7 @@
+mask_aggregation: aggregate masked field values
+-----------------------------------------------
+
+.. automodule:: umami.calculations.metric.mask_aggregation
+   :members:
+   :undoc-members:
+   :show-inheritance:

docs/source/umami.calculations.rst

@@ -12,6 +12,7 @@ Calculations for Metrics or Residuals
    umami.calculations.metric.chi_intercept_gradient
    umami.calculations.metric.count_equal
    umami.calculations.metric.hypsometric_integral
+   umami.calculations.metric.mask_aggregation
    umami.calculations.metric.watershed_aggregation
 
 Calculations for Residuals only

umami/calculations/__init__.py

@@ -11,6 +11,7 @@
     count_equal,
     hypsometric_integral,
     watershed_aggregation,
+    mask_aggregation,
 )
 from .residual import (
     discretized_misfit,
@@ -26,6 +27,7 @@
     "count_equal",
     "hypsometric_integral",
     "watershed_aggregation",
+    "mask_aggregation",
     "discretized_misfit",
     "joint_density_misfit",
     "kstest",

umami/calculations/metric/__init__.py

@@ -3,6 +3,7 @@
 from .count_equal import count_equal
 from .hypsometric_integral import hypsometric_integral
 from .watershed_aggregation import watershed_aggregation
+from .mask_aggregation import mask_aggregation
 
 __all__ = [
     "aggregate",
@@ -11,4 +12,5 @@
     "count_equal",
     "hypsometric_integral",
     "watershed_aggregation",
+    "mask_aggregation",
 ]

umami/calculations/metric/mask_aggregation.py

@@ -0,0 +1,94 @@
+import numpy as np
+
+from .aggregate import _aggregate
+
+
+def mask_aggregation(grid, field, mask, method, **kwds):
+    """Aggregate a field value masked by a boolean field.
+
+    ``mask_aggregation`` calculates aggregate values on a field masked by a
+    second, boolean field. It supports all methods in the `numpy`_ namespace
+    that reduce an array to a scalar.
+
+    .. _numpy: https://numpy.org
+
+    Parameters
+    ----------
+    grid : Landlab model grid
+    field : str
+        An at-node Landlab grid field that is present on the model grid.
+    mask : str
+        An at-node Landlab grid field of boolean type. Aggregation is done
+        where the masked value is True.
+    method : str
+        The name of a numpy namespace method.
+    **kwds
+        Any additional keyword arguments needed by the method.
+
+    Returns
+    -------
+    out : float
+        The aggregate value.
+
+    Examples
+    --------
+    First an example that only uses the ``mask_aggregation`` function.
+
+    >>> from landlab import RasterModelGrid
+    >>> from landlab.components import FlowAccumulator
+    >>> from umami.calculations import mask_aggregation
+    >>> grid = RasterModelGrid((10, 10))
+    >>> z = grid.add_zeros("node", "topographic__elevation")
+    >>> z += grid.x_of_node + grid.y_of_node
+
+    Create a boolean mask and add it to the grid.
+
+    >>> mask = grid.add_field("mask", z > 11, at="node")
+
+    ``mask_aggregation`` supports all functions in the `numpy`_ namespace.
+    Here we show `mean`_ and `percentile`_. The latter of which takes an
+    additional argument, *q*.
+
+    .. _numpy: https://numpy.org
+    .. _mean: https://docs.scipy.org/doc/numpy/reference/generated/numpy.mean.html
+    .. _percentile: https://docs.scipy.org/doc/numpy/reference/generated/numpy.percentile.html
+
+
+    >>> mask_aggregation(grid, "topographic__elevation", "mask",  "mean")
+    14.0
+    >>> mask_aggregation(
+    ... grid,
+    ... "topographic__elevation",
+    ... "mask",
+    ... "percentile",
+    ... q=10)
+    12.0
+
+    Next, the same calculations are shown as part of an umami ``Metric``.
+
+    >>> from io import StringIO
+    >>> from umami import Metric
+    >>> file_like=StringIO('''
+    ... mask_mean:
+    ...     _func: mask_aggregation
+    ...     mask: mask
+    ...     method: mean
+    ...     field: topographic__elevation
+    ... mask_10thptile:
+    ...     _func: mask_aggregation
+    ...     mask: mask
+    ...     method: percentile
+    ...     field: topographic__elevation
+    ...     q: 10
+    ... ''')
+    >>> metric = Metric(grid)
+    >>> metric.add_from_file(file_like)
+    >>> metric.names
+    ['mask_mean', 'mask_10thptile']
+    >>> metric.calculate()
+    >>> metric.values
+    [14.0, 12.0]
+    """
+    masked = grid.at_node[mask]
+    vals = grid.at_node[field][masked]
+    return _aggregate(vals, method, **kwds)