Merge pull request #149 from NOAA-OWP/public_dataformat

Public rasterize and vectorize method

pyproject.toml

@@ -12,7 +12,7 @@ authors = [
 requires-python = ">=3.8"
 keywords = ["geospatial", "evaluations"]
 license = {text = "MIT"}
-version = "0.1.1"
+version = "0.1.2"
 dynamic = ["readme", "dependencies"]
 
 

src/gval/accessors/gval_dataframe.py

@@ -3,9 +3,11 @@
 
 import pandas as pd
 from pandera.typing import DataFrame
+import xarray as xr
 
 from gval.comparison.compute_categorical_metrics import _compute_categorical_metrics
 from gval.utils.schemas import Metrics_df
+from gval.homogenize.rasterize import _rasterize_data
 
 
 @pd.api.extensions.register_dataframe_accessor("gval")
@@ -88,3 +90,38 @@ def compute_categorical_metrics(
             average=average,
             weights=weights,
         )
+
+    def rasterize_data(
+        self, reference_map: Union[xr.Dataset, xr.DataArray], rasterize_attributes: list
+    ) -> Union[xr.Dataset, xr.DataArray]:
+        """
+        Convenience function for rasterizing vector data using a reference raster.  For more control use `make_geocube`
+        from the geocube package.
+
+        Parameters
+        ----------
+        reference_map: Union[xr.Dataset, xr.DataArray]
+            Map to reference in creation of rasterized vector map
+        rasterize_attributes: list
+            Attributes to rasterize
+
+        Returns
+        -------
+        Union[xr.Dataset, xr.DataArray]
+            Rasterized Data
+
+        Raises
+        ------
+        KeyError
+
+        References
+        ----------
+        .. [1] [geocube `make_geocube`](https://corteva.github.io/geocube/html/geocube.html)
+
+        """
+
+        return _rasterize_data(
+            candidate_map=reference_map,
+            benchmark_map=self._obj,
+            rasterize_attributes=rasterize_attributes,
+        )

src/gval/accessors/gval_xarray.py

@@ -462,7 +462,7 @@ def compute_agreement_map(
         )
 
         if self.agreement_map_format == "vector":
-            agreement_map = _vectorize_data(agreement_map)
+            agreement_map = agreement_map.gval.vectorize_data()
 
         return agreement_map
 
@@ -677,3 +677,15 @@ def cont_plot(
             basemap=basemap,
             colorbar_label=colorbar_label,
         )
+
+    def vectorize_data(self) -> gpd.GeoDataFrame:
+        """
+        Vectorize an xarray DataArray or Dataset
+
+        Returns
+        -------
+        gpd.GeoDataFrame
+            Vectorized data
+        """
+
+        return _vectorize_data(self._obj)

src/gval/comparison/compute_comparison.py

@@ -284,7 +284,7 @@ def wrapper(*args, **kwargs):
                         unique_benchmark_values=kwargs["allow_benchmark_values"],
                     )
 
-            else:
+            if "comparison_function" not in kwargs:
                 kwargs["comparison_function"] = getattr(self, "szudzik")
 
             # Call the decorated function

src/gval/statistics/categorical_stat_funcs.py

@@ -324,7 +324,7 @@ def prevalence(tp: Number, tn: Number, fp: Number, fn: Number) -> float:
     ----------
     .. [1] [Prevalence](https://en.wikipedia.org/wiki/Prevalence)
     """
-    return (tp + fp) / (tp + fp + tn + fn)
+    return (tp + fn) / (tp + fp + tn + fn)
 
 
 def accuracy(tp: Number, tn: Number, fp: Number, fn: Number) -> float:

tests/cases_accessors.py

@@ -391,6 +391,14 @@ def case_continuous_plot_fail(candidate_map):
     return candidate_map
 
 
+@parametrize(
+    "vector_map, reference_map, attributes",
+    list(zip(benchmark_maps[2:3], candidate_maps[0:1], [["category"]])),
+)
+def case_dataframe_accessor_rasterize(vector_map, reference_map, attributes):
+    return vector_map, reference_map, attributes
+
+
 candidate_maps = ["candidate_continuous_0.tif", "candidate_continuous_1.tif"]
 benchmark_maps = ["benchmark_continuous_0.tif", "benchmark_continuous_1.tif"]
 
@@ -424,4 +432,4 @@ def case_data_set_accessor_continuous(candidate_map, benchmark_map):
     list(zip(candidate_maps, benchmark_maps, agreement_maps)),
 )
 def case_accessor_attributes(candidate_map, benchmark_map, agreement_map):
-    return (_load_xarray(candidate_map), _load_xarray(benchmark_map), agreement_map)
+    return _load_xarray(candidate_map), _load_xarray(benchmark_map), agreement_map

tests/cases_compute_categorical_metrics.py

@@ -35,7 +35,7 @@
             "overall_bias": {0: 0.9535906369448297},
             "positive_likelihood_ratio": {0: 1.3660848657244327},
             "positive_predictive_value": {0: 0.7891880387357667},
-            "prevalence": {0: 0.6986443954301008},
+            "prevalence": {0: 0.7326460310772968},
             "prevalence_threshold": {0: 0.46108525048654536},
             "true_negative_rate": {0: 0.44911012235817577},
             "true_positive_rate": {0: 0.7525623245272807},
@@ -85,7 +85,7 @@
                 0: 0.31210513014910285,
                 1: 0.6623376623376623,
             },
-            "prevalence": {0: 0.42776066158586024, 1: 0.4514317452480869},
+            "prevalence": {0: 0.1928544403005243, 1: 0.5266600839298938},
             "prevalence_threshold": {0: 0.4205207112769604, 1: 0.42959744253992793},
             "true_negative_rate": {0: 0.635438291033282, 1: 0.6779661016949152},
             "true_positive_rate": {0: 0.6922645739910314, 1: 0.5677290836653387},
@@ -476,12 +476,12 @@ def case_compute_categorical_metrics_fail(
                 "tn": [28, 28, 20, 20, 12, 12],
                 "tp": [1, 1, 5, 5, 9, 9],
                 "prevalence": [
-                    0.133333,
-                    0.133333,
-                    0.333333,
-                    0.333333,
-                    0.533333,
-                    0.533333,
+                    0.26666666666666666,
+                    0.26666666666666666,
+                    0.3333333333333333,
+                    0.3333333333333333,
+                    0.4,
+                    0.4,
                 ],
                 "true_negative_rate": [
                     0.848485,

tests/test_accessors.py

@@ -317,3 +317,16 @@ def test_accessor_attributes(candidate_map, benchmark_map, agreement_map):
 
     assert isinstance(agreement_map, xr.DataArray)
     assert isinstance(attrs_df, DataFrame)
+
+
+@parametrize_with_cases(
+    "vector_map, reference_map, attributes",
+    glob="dataframe_accessor_rasterize",
+)
+def test_dataframe_accessor_rasterize(vector_map, reference_map, attributes):
+    raster_map = vector_map.gval.rasterize_data(
+        reference_map=reference_map, rasterize_attributes=attributes
+    )
+
+    assert isinstance(raster_map, type(reference_map))
+    assert raster_map.shape == reference_map.shape