lsapy 0.1.0.dev0__py3-none-any.whl → 0.1.0.dev1__py3-none-any.whl

lsapy/__init__.py

@@ -1,9 +1,9 @@
 """A Python package for Land Suitability Analysis."""
 
-from lsapy.lsa import LandSuitability
 from lsapy.criteria import SuitabilityCriteria
 from lsapy.functions import *
+from lsapy.lsa import LandSuitability
 
 __author__ = "Baptiste Hamon"
 __email__ = "baptiste.hamon@pg.canterbury.ac.nz"
-__version__ = "0.1.0-dev0"
+__version__ = "0.1.0-dev1"

lsapy/criteria.py

@@ -1,22 +1,22 @@
 """Suitability Criteria definition."""
 
-from typing import Optional, Union, Any
 import xarray as xr
 
-from lsapy.functions import SuitabilityFunction, MembershipSuitFunction, DiscreteSuitFunction
+from lsapy.functions import DiscreteSuitFunction, MembershipSuitFunction, SuitabilityFunction
 
 __all__ = ["SuitabilityCriteria"]
 
+
 class SuitabilityCriteria:
     def __init__(
             self,
             name: str,
             indicator: xr.Dataset | xr.DataArray,
-            func: Union[SuitabilityFunction, MembershipSuitFunction, DiscreteSuitFunction],
-            weight: Optional[Union[int,float]] = 1,
-            category: Optional[str] = None,
-            long_name: Optional[str] = None,
-            description: Optional[str] = None
+            func: SuitabilityFunction | MembershipSuitFunction | DiscreteSuitFunction,
+            weight: int | float | None = 1,
+            category: str | None = None,
+            long_name: str | None = None,
+            description: str | None = None
 
     ) -> None:
         self.name = name
@@ -27,7 +27,7 @@ class SuitabilityCriteria:
         self.long_name = long_name
         self.description = description
         self._from_indicator = _get_indicator_description(indicator)
-    
+
     def __repr__(self) -> str:
         attrs = []
         attrs.append(f"name='{self.name}'")
@@ -41,17 +41,20 @@ class SuitabilityCriteria:
         if self.description is not None:
             attrs.append(f"description='{self.description}'")
         return f"{self.__class__.__name__}({', '.join(attrs) if attrs else ''})"
-        
-    
+
     def compute(self) -> xr.DataArray:
-        if self.func.func_method == 'discrete': # need to vectorize the discrete function
-            sc : xr.DataArray = xr.apply_ufunc(self.func.map, self.indicator).rename(self.name)
+        if self.func.func_method == 'discrete':  # need to vectorize the discrete function
+            sc: xr.DataArray = xr.apply_ufunc(self.func.map, self.indicator).rename(self.name)
         else:
-            sc : xr.DataArray = self.func.map(self.indicator).rename(self.name)
-        return sc.assign_attrs(dict({k: v for k, v in self.attrs.items() if k not in ['name', 'func_method', 'from_indicator']},
-                                    **{'history': f"func_method: {self.func}; from_indicator: [{self._from_indicator}]", 'compute': 'done'}))
-    
-    
+            sc: xr.DataArray = self.func.map(self.indicator).rename(self.name)
+        return sc.assign_attrs(
+            dict({k: v for k, v in self.attrs.items() if k not in ['name', 'func_method', 'from_indicator']},
+                 **{
+                     'history': f"func_method: {self.func}; from_indicator: [{self._from_indicator}]",
+                     'compute': 'done'
+                 })
+        )
+
     @property
     def attrs(self):
         return {k: v for k, v in {
@@ -69,4 +72,4 @@ def _get_indicator_description(indicator: xr.Dataset | xr.DataArray) -> str:
     if indicator.attrs != {}:
         return f"name: {indicator.name}; " + "; ".join([f"{k}: {v}" for k, v in indicator.attrs.items()])
     else:
-        return f"name: {indicator.name}"
+        return f"name: {indicator.name}"

lsapy/functions.py

@@ -1,13 +1,13 @@
 """Suitability Functions definitions."""
 
-from typing import Optional, Callable, Union, Any
 import warnings
+from collections.abc import Callable
+from typing import Any
+
 import matplotlib.pyplot as plt
 import numpy as np
-import xarray as xr
 from scipy.optimize import curve_fit
 
-
 __all__ = [
     "SuitabilityFunction",
     "MembershipSuitFunction",
@@ -19,42 +19,42 @@ class SuitabilityFunction:
 
     def __init__(
             self,
-            func: Optional[Callable] = None,
-            func_method: Optional[str] = None,
-            func_params: Union[dict[str, Any]] = {}
+            func: Callable | None = None,
+            func_method: str | None = None,
+            func_params: dict[str, Any] = None
     ):
         if func_params is not None:
             if func is None and func_method is None:
                 raise ValueError("If `func_params` is provided, `func` or `func_method` must also be provided.")
+        else:
+            func_params = {}
 
         self.func = func
         self.func_method = func_method
         self.func_params = func_params
         if func is None and func_method is not None:
             self.func = _get_function_from_name(func_method)
-    
 
     def __repr__(self):
-        return f"{self.__class__.__name__}(func={self.func.__name__}, func_method='{self.func_method}', func_params={self.func_params})"
-    
+        return (f"{self.__class__.__name__}("
+                f"func={self.func.__name__}, "
+                f"func_method='{self.func_method}', "
+                f"func_params={self.func_params})")
 
     def __call__(self, x):
         if self.func is None:
             raise ValueError("No function has been provided.")
         return self.func(x, **self.func_params)
-    
-    
+
     def map(self, x):
         return self(x)
-    
 
     def plot(self, x) -> None:
         plt.plot(x, self(x))
 
-
     @property
     def attrs(self):
-        if self.func_method == None and self.func_params == None:
+        if self.func_method is None and self.func_params is None:
             return {}
         return {k: v for k, v in {
                     'func_method': self.func_method,
@@ -71,16 +71,17 @@ class MembershipSuitFunction(SuitabilityFunction):
 
     def __init__(
             self,
-            func: Optional[Callable] = None,
-            func_method: Optional[str] = None,
-            func_params: Optional[dict[str, int | float]] = None
+            func: Callable | None = None,
+            func_method: str | None = None,
+            func_params: dict[str, int | float] | None = None
     ):
         super().__init__(func, func_method, func_params)
 
-
     @staticmethod
-    def fit(x, y = np.array([0,.25,.5,.75,1]), methods: str | list[str] = 'all', plot: bool = False):
-        return _fit_mbs_functions(x, y, methods, plot)
+    def fit(x, y=None, methods: str | list[str] = 'all', plot: bool = False):
+        if y is None:
+            y = [0, .25, .5, .75, 1]
+        return _fit_mbs_functions(x, np.array(y), methods, plot)
 
 
 def _prepare_for_fitting(methods:  str | list[str] = 'all'):
@@ -92,16 +93,16 @@ def _prepare_for_fitting(methods:  str | list[str] = 'all'):
     elif isinstance(methods, list) or isinstance(methods, str):
         if isinstance(methods, str):
             methods = [methods]
-        
+
         _methods = []
         for method in methods:
             if method in _types:
                 [_methods.append(m) for m in equations[method.replace('_like', '')].keys()]
             else:
-                try :
+                try:
                     _get_function_from_name(method)
                     _methods.append(method)
-                except :
+                except Exception:
                     _skipped.append(method)
                     warnings.warn(f"`{method}` not found in equations. Skipped.", stacklevel=2)
         methods = _methods
@@ -125,7 +126,7 @@ def _get_function_p0(method: str, x: np.ndarray) -> list[float]:
 
 
 def _fit_mbs_functions(x, y, methods: str | list[str] = 'all', plot: bool = False):
-    
+
     skipped = []
     methods, _skipped = _prepare_for_fitting(methods)
     skipped.extend(_skipped)
@@ -133,22 +134,22 @@ def _fit_mbs_functions(x, y, methods: str | list[str] = 'all', plot: bool = Fals
     if len(methods) == 0:
         print(f"Skipped fitting for the following methods: {', '.join(skipped)}.")
         raise ValueError("No methods to fit.")
-    else :
+    else:
         x_ = np.linspace(min(x), max(x), 100)
         rms_errors = []
         f_params = []
         for method in methods:
-            try :
+            try:
                 f = _get_function_from_name(method)
                 p0 = _get_function_p0(method, x)
                 popt, _ = curve_fit(f, x, y, p0=p0, maxfev=15000)
                 y_ = f(x_, *popt)
                 f_params.append(popt)
-                rmse = _rms_error(y,f(x, *popt))
+                rmse = _rms_error(y, f(x, *popt))
                 rms_errors.append(rmse)
                 if plot:
                     plt.plot(x_, y_, label=method + f' (RMSE={rmse:.2f})')
-            except:
+            except Exception:
                 skipped.append(method)
                 warnings.warn(f"Failed to fit `{method}`. Skipped.", stacklevel=2)
         if plot:
@@ -158,7 +159,7 @@ def _fit_mbs_functions(x, y, methods: str | list[str] = 'all', plot: bool = Fals
 
         if len(skipped) > 0:
             print(f"Skipped fitting for the following methods: {', '.join(skipped)}.")
-    f_best, p_best = _get_best_fit(methods, rms_errors, f_params)
+    f_best, p_best = _get_best_fit([m for m in methods if m not in skipped], rms_errors, f_params)
     return _get_function_from_name(f_best), p_best
 
 
@@ -170,7 +171,7 @@ class DiscreteSuitFunction(SuitabilityFunction):
 
     def __init__(
             self,
-            func_params: Optional[dict[str, int | float]] = None
+            func_params: dict[str, int | float] | None = None
     ):
         self.func = discrete
         self.func_method = 'discrete'
@@ -181,7 +182,7 @@ class DiscreteSuitFunction(SuitabilityFunction):
 # ---------------------------- Utility functions ----------------------------- #
 # ---------------------------------------------------------------------------- #
 
-equations : dict[str, dict] = {}
+equations: dict[str, dict] = {}
 
 
 def _get_function_from_name(name: str) -> callable:
@@ -199,7 +200,7 @@ def equation(type: str):
     ----------
     type : str
         The type of equation to register.
-    
+
     Returns
     -------
     decorator
@@ -216,13 +217,13 @@ def equation(type: str):
 
 
 @equation('discrete')
-def discrete(x, rules: dict[str|int, int|float]) -> float:
+def discrete(x, rules: dict[str | int, int | float]) -> float:
     return np.vectorize(rules.get, otypes=[np.float32])(x, np.nan)
 
 
 @equation('sigmoid')
 def logistic(x, a, b):
-    return 1 / (1 + np.exp(-a*(x - b)))
+    return 1 / (1 + np.exp(-a * (x - b)))
 
 
 @equation('sigmoid')

lsapy/lsa.py

@@ -1,29 +1,27 @@
 """Land Suitability definition."""
 
-from typing import Optional, Any, Union
-import xarray as xr
+from typing import Any
+
+import geopandas as gpd
 import numpy as np
 import pandas as pd
-import geopandas as gpd
+import xarray as xr
 from shapely.geometry import mapping
-import rioxarray
-import regionmask
 
 from lsapy.criteria import SuitabilityCriteria
-from lsapy.statistics import statistics_summary, spatial_statistics_summary
+from lsapy.statistics import spatial_statistics_summary, statistics_summary
 
 __all__ = ["LandSuitability"]
 
 
 class LandSuitability:
-    
     def __init__(
-            self,
-            name: str,
-            criteria: dict[str, SuitabilityCriteria],
-            short_name: Optional[str] = None,
-            long_name: Optional[str] = None,
-            description: Optional[str] = None,
+        self,
+        name: str,
+        criteria: dict[str, SuitabilityCriteria],
+        short_name: str | None = None,
+        long_name: str | None = None,
+        description: str | None = None,
     ) -> None:
         self.name = name
         self.criteria = criteria
@@ -31,48 +29,50 @@ class LandSuitability:
         self.long_name = long_name
         self.description = description
 
-        self._sort_criteria_by_weight() # important if suitability as limited factor: return most weighted criteria if equals
+        self._sort_criteria_by_weight()  # important if suitability as limited factor
         self._criteria_name_list = [sc.name for sc in self.criteria.values()]
         self._category_list = list(set([sc.category for sc in self.criteria.values()]))
         self._get_params_by_category()
 
-
     def __repr__(self) -> str:
-        if hasattr(self, 'data'):
+        if hasattr(self, "data"):
             return self.data.__repr__()
         else:
             attrs = []
             for k, v in self.attrs.items():
                 if isinstance(v, str):
-                    v = f"'{v}'"
-                attrs.append(f"{k}={v}")
+                    v_ = f"'{v}'"
+                else:
+                    v_ = v
+                attrs.append(f"{k}={v_}")
             return f"{self.__class__.__name__}({', '.join(attrs) if attrs else ''})"
-    
 
     @property
     def attrs(self):
-        return {k: v for k, v in {
-                    'name': self.name,
-                    'criteria': self._criteria_name_list,
-                    'short_name': self.short_name,
-                    'long_name': self.long_name,
-                    'description': self.description,
-                }.items() if v is not None}
-
+        return {
+            k: v
+            for k, v in {
+                "name": self.name,
+                "criteria": self._criteria_name_list,
+                "short_name": self.short_name,
+                "long_name": self.long_name,
+                "description": self.description,
+            }.items()
+            if v is not None
+        }
 
     # def __getitem__(self, key: str) -> SuitabilityCriteria:
     #     return self.criteria[key]
 
-
-    def compute_criteria_suitability(self, inplace: Optional[bool] = False) -> None | xr.Dataset:
+    def compute_criteria_suitability(self, inplace: bool | None = False) -> None | xr.Dataset:
         sc_list = []
         for _, sc in self.criteria.items():
-            print(f'Computing {sc.name}...')
-            if sc.indicator.attrs.get('compute', '') == 'done':
+            print(f"Computing {sc.name}...")
+            if sc.indicator.attrs.get("compute", "") == "done":
                 sc_list.append(sc.indicator.rename(sc.name))
             else:
                 sc_list.append(sc.compute())
-        ls = xr.merge(sc_list, compat='override', combine_attrs='drop')
+        ls = xr.merge(sc_list, compat="override", combine_attrs="drop")
         for sc in sc_list:
             ls[sc.name].attrs = sc.attrs
         ls.attrs = self.attrs
@@ -80,173 +80,205 @@ class LandSuitability:
             self.data = ls
         else:
             return ls
-    
 
     def compute_category_suitability(
-            self, method: str,
-            keep_criteria: Optional[bool] = False,
-            inplace: Optional[bool] = False,
-            limit_var: Optional[bool] = False,
+        self,
+        method: str,
+        keep_criteria: bool | None = False,
+        inplace: bool | None = False,
+        limit_var: bool | None = False,
     ) -> xr.Dataset:
-        if not hasattr(self, 'data'):
+        if not hasattr(self, "data"):
             ds = self.compute_criteria_suitability()
         else:
             ds = self.data
-        
+
         out = []
         out_attrs = []
         for category in self._category_list:
-            print(f'Computing {category}...')
+            print(f"Computing {category}...")
             sc_list = [sc for sc in self.criteria.values() if sc.category == category]
-            res = _aggregate_vars(ds[self._criteria_by_category[category]], method=method,
-                                  weights=[sc.weight for sc in sc_list], limit_var=limit_var)
+            res = _aggregate_vars(
+                ds[self._criteria_by_category[category]],
+                method=method,
+                weights=[sc.weight for sc in sc_list],
+                limit_var=limit_var,
+            )
             if isinstance(res, xr.Dataset):
-                res = res.rename({'limiting_factor': f'{category}', 'limiting_var': f'{category}_var'})
+                res = res.rename({"limiting_factor": f"{category}", "limiting_var": f"{category}_var"})
             else:
-                res = res.rename(f'{category}_suitability')
-            res.attrs.update({'long_name': f'{category.capitalize()} Suitability'})
+                res = res.rename(f"{category}_suitability")
+            res.attrs.update({"long_name": f"{category.capitalize()} Suitability"})
             out.append(res)
             out_attrs.append(res.attrs)
-        out = xr.merge(out, compat='override', combine_attrs='drop')
+        out = xr.merge(out, compat="override", combine_attrs="drop")
         if keep_criteria:
-            out = xr.merge([ds, out], compat='override', combine_attrs='drop')
-            for sc in ds.data_vars: # add attributes of criteria
+            out = xr.merge([ds, out], compat="override", combine_attrs="drop")
+            for sc in ds.data_vars:  # add attributes of criteria
                 out[sc].attrs = ds[sc].attrs
-        for i, category in enumerate(self._category_list): # add attributes of category suitability
-            out[f'{category}_suitability'].attrs = out_attrs[i]
-        
+        for i, category in enumerate(self._category_list):  # add attributes of category suitability
+            out[f"{category}_suitability"].attrs = out_attrs[i]
+
         out.attrs.update(self.attrs)
         if inplace:
             self.data = out
         else:
             return out
-    
-
-    def compute_suitability(self,
-                            method: Union[str, dict[str,str]] = 'mean',
-                            by_category: Optional[bool] = False,
-                            keep_all: Optional[bool] = False,
-                            inplace: Optional[bool] = False,
-                            limit_var: Optional[bool] = False
+
+    def compute_suitability(
+        self,
+        method: str | dict[str, str] = "mean",
+        by_category: bool | None = False,
+        keep_all: bool | None = False,
+        inplace: bool | None = False,
+        limit_var: bool | None = False,
     ) -> xr.Dataset:
         if isinstance(method, str):
             cat_method, suit_method = method, method
         elif isinstance(method, dict):
-            cat_method = method.get('category', 'mean')
-            suit_method = method.get('overall', 'mean')
+            cat_method = method.get("category", "mean")
+            suit_method = method.get("overall", "mean")
         else:
             raise ValueError("Method must be a string or a dictionary.")
-        
-        if not hasattr(self, 'data'):
+
+        if not hasattr(self, "data"):
             if by_category:
-                ds = self.compute_category_suitability(method=cat_method, keep_criteria=True, inplace=False, limit_var=limit_var)
+                ds = self.compute_category_suitability(
+                    method=cat_method, keep_criteria=True, inplace=False, limit_var=limit_var
+                )
             else:
                 ds = self.compute_criteria_suitability(inplace=False)
         else:
             ds = self.data
-        
+
         if by_category:
             weights = [self.weights_by_category[category] for category in self._category_list]
-            on_vars = [f'{category}_suitability' for category in self._category_list]
-        else :
+            on_vars = [f"{category}_suitability" for category in self._category_list]
+        else:
             weights = [sc.weight for sc in self.criteria.values()]
             on_vars = self._criteria_name_list
-        
-        print(f'Computing suitability...')
-        out = _aggregate_vars(ds[on_vars], method=suit_method, weights=weights, limit_var=limit_var).rename('suitability')
-        out.attrs.update({'long_name': 'Suitability'})
+
+        print("Computing suitability...")
+        out = _aggregate_vars(ds[on_vars], method=suit_method, weights=weights, limit_var=limit_var).rename(
+            "suitability"
+        )
+        out.attrs.update({"long_name": "Suitability"})
         out_attrs = out.attrs
 
         if keep_all:
-            out = xr.merge([ds, out], compat='override', combine_attrs='drop')
-            for sc in ds.data_vars: # add attributes of criteria
+            out = xr.merge([ds, out], compat="override", combine_attrs="drop")
+            for sc in ds.data_vars:  # add attributes of criteria
                 out[sc].attrs = ds[sc].attrs
-            out['suitability'].attrs = out_attrs
-        
+            out["suitability"].attrs = out_attrs
+
         out.attrs.update(self.attrs)
         if inplace:
             self.data = out
         else:
             return out
-    
 
     def _sort_criteria_by_weight(self) -> dict[str, SuitabilityCriteria]:
         self.criteria = dict(sorted(self.criteria.items(), key=lambda item: item[1].weight, reverse=True))
 
-
     def _get_params_by_category(self):
         self._get_criteria_by_category()
         self._get_weights_by_category()
-    
-    
+
     def _get_criteria_by_category(self) -> dict[str, list[str]]:
         self._criteria_by_category = {category: [] for category in self._category_list}
         for sc in self.criteria.values():
             self._criteria_by_category[sc.category].append(sc.name)
-    
 
     def _get_weights_by_category(self) -> dict[str, float | int]:
         self.weights_by_category = {category: [] for category in self._category_list}
         for category in self._category_list:
-            self.weights_by_category[category] = sum([sc.weight for sc in self.criteria.values() if sc.category == category])
-
-    
-    def mask(self, mask : Union[xr.DataArray, gpd.GeoDataFrame], inplace : Optional[bool] = False,
-          spatial_dims : Optional[tuple[str, str]] = None, crs : Optional[str] = None, invert : bool = False, **kwargs) -> xr.DataArray | xr.Dataset:
-        
-        if not hasattr(self, 'data'):
+            self.weights_by_category[category] = sum(
+                [sc.weight for sc in self.criteria.values() if sc.category == category]
+            )
+
+    def mask(
+        self,
+        mask: xr.DataArray | gpd.GeoDataFrame,
+        inplace: bool | None = False,
+        spatial_dims: tuple[str, str] | None = None,
+        crs: str | None = None,
+        invert: bool = False,
+        **kwargs,
+    ) -> xr.DataArray | xr.Dataset:
+        if not hasattr(self, "data"):
             raise ValueError("Suitability must be computed first.")
-        
+
         if inplace:
             self.data = _mask_data(self.data, mask, spatial_dims=spatial_dims, crs=crs, invert=invert, **kwargs)
-        else:     
+        else:
             return _mask_data(self.data, mask, spatial_dims=spatial_dims, crs=crs, invert=invert, **kwargs)
-    
 
     def statistics(
-            self,
-            on_vars: Optional[list] = None,
-            on_dims: Optional[list] = None,
-            on_dim_values: Optional[dict[str, Any]] = None,
-            bins : Optional[Union[list, np.ndarray]] = None,
-            bins_labels : Optional[list] = None,
-            all_bins : Optional[bool] = False,
-            cell_area : Optional[tuple[Union[float, str], str]] = None,
-            dropna: Optional[bool] = False,
-            **kwargs) -> pd.DataFrame:
-        
-        if not hasattr(self, 'data'):
+        self,
+        on_vars: list | None = None,
+        on_dims: list | None = None,
+        on_dim_values: dict[str, Any] | None = None,
+        bins: list | np.ndarray | None = None,
+        bins_labels: list | None = None,
+        all_bins: bool | None = False,
+        cell_area: tuple[float | str, str] | None = None,
+        dropna: bool | None = False,
+        **kwargs,
+    ) -> pd.DataFrame:
+        if not hasattr(self, "data"):
             raise ValueError("Suitability must be computed first.")
-        
-        return statistics_summary(self.data, on_vars=on_vars, on_dims=on_dims, on_dim_values=on_dim_values,
-                                  bins=bins, bins_labels=bins_labels, all_bins=all_bins,
-                                  cell_area=cell_area, dropna=dropna **kwargs)
-    
+
+        return statistics_summary(
+            self.data,
+            on_vars=on_vars,
+            on_dims=on_dims,
+            on_dim_values=on_dim_values,
+            bins=bins,
+            bins_labels=bins_labels,
+            all_bins=all_bins,
+            cell_area=cell_area,
+            dropna=dropna**kwargs,
+        )
 
     def spatial_statistics(
-            self,
-            areas : gpd.GeoDataFrame,
-            name : Optional[str] = 'area',
-            on_vars: Optional[list] = None,
-            on_dims: Optional[list] = None,
-            on_dim_values: Optional[dict[str, Any]] = None,
-            bins : Optional[np.ndarray] = None,
-            all_bins : Optional[bool] = False,
-            cell_area : Optional[tuple[Union[float, str], str]] = None,
-            mask_kwargs : dict = {},
-            stats_kwargs: dict = {}) -> pd.DataFrame:
-        
-        if not hasattr(self, 'data'):
+        self,
+        areas: gpd.GeoDataFrame,
+        name: str | None = "area",
+        on_vars: list | None = None,
+        on_dims: list | None = None,
+        on_dim_values: dict[str, Any] | None = None,
+        bins: np.ndarray | None = None,
+        all_bins: bool | None = False,
+        cell_area: tuple[float | str, str] | None = None,
+        mask_kwargs: dict = None,
+        stats_kwargs: dict = None,
+    ) -> pd.DataFrame:
+        if not hasattr(self, "data"):
             raise ValueError("Suitability must be computed first.")
-        
-        return spatial_statistics_summary(self.data, areas, name=name, on_vars=on_vars, on_dims=on_dims, on_dim_values=on_dim_values,
-                                   bins=bins, all_bins=all_bins, cell_area=cell_area, mask_kwargs=mask_kwargs, stats_kwargs=stats_kwargs)
-
 
-def _mask_data(data : xr.DataArray | xr.Dataset, mask : Union[xr.DataArray, gpd.GeoDataFrame],
-          spatial_dims : Optional[tuple[str, str]] = None,
-          crs : Optional[str] = None, invert : bool = False, **kwargs) -> xr.DataArray | xr.Dataset:
-    
+        return spatial_statistics_summary(
+            self.data,
+            areas,
+            name=name,
+            on_vars=on_vars,
+            on_dims=on_dims,
+            on_dim_values=on_dim_values,
+            bins=bins,
+            all_bins=all_bins,
+            cell_area=cell_area,
+            mask_kwargs=mask_kwargs,
+            stats_kwargs=stats_kwargs,
+        )
+
+
+def _mask_data(
+    data: xr.DataArray | xr.Dataset,
+    mask: xr.DataArray | gpd.GeoDataFrame,
+    spatial_dims: tuple[str, str] | None = None,
+    crs: str | None = None,
+    invert: bool = False,
+    **kwargs,
+) -> xr.DataArray | xr.Dataset:
     if isinstance(mask, gpd.GeoDataFrame):
         mask = mask.to_crs(crs)
         data = data.rio.set_spatial_dims(*spatial_dims).rio.write_crs(crs)
@@ -256,76 +288,102 @@ def _mask_data(data : xr.DataArray | xr.Dataset, mask : Union[xr.DataArray, gpd.
             mask = ~mask
         return data.where(mask, **kwargs)
     else:
-        raise ValueError('mask must be a GeoDataFrame or DataArray')
+        raise ValueError("mask must be a GeoDataFrame or DataArray")
 
 
 ####################################################################################################
 # VARIABLES AGGREGATION FUNCTIONS
 ####################################################################################################
 
-def vars_weighted_mean(ds: xr.Dataset, vars = None, weights = None) -> xr.DataArray:
+
+def vars_weighted_mean(ds: xr.Dataset, vars=None, weights=None) -> xr.DataArray:
     if vars is None:
         vars = list(ds.data_vars)
     if weights is None:
         weights = np.ones(len(vars))
-    
-    s = sum([ds[v] * w for v, w in zip(vars, weights)])
-    da: xr.DataArray = s / sum(weights)
-    return da.assign_attrs({
-        'method': 'Weighted Mean', 
-        'description': f"Weighted Mean of variables: {', '.join([f'{v} ({w})' for v, w in zip(vars, weights)])}."
-    }).rename('weighted_mean')
 
-def vars_mean(ds: xr.Dataset, vars = None) -> xr.DataArray:
-    if vars is None: vars = list(ds.data_vars)
+    s = sum([ds[v] * w for v, w in zip(vars, weights, strict=False)])
+    da: xr.DataArray = s / sum(weights)
+    return da.assign_attrs(
+        {
+            "method": "Weighted Mean",
+            "descritpion": (
+                "Weighted Mean of variables: "
+                f"{', '.join([f'{v} ({w})' for v, w in zip(vars, weights, strict=False)])}."
+            )
+        }
+    ).rename("weighted_mean")
+
+
+def vars_mean(ds: xr.Dataset, vars=None) -> xr.DataArray:
+    if vars is None:
+        vars = list(ds.data_vars)
     da = vars_weighted_mean(ds, vars=vars)
-    return da.assign_attrs({'method': 'Mean', 'description': f"Mean of variables: {', '.join(vars)}."}).rename('mean')
+    return da.assign_attrs({"method": "Mean", "description": f"Mean of variables: {', '.join(vars)}."}).rename("mean")
+
 
-def vars_weighted_geomean(ds: xr.Dataset, vars = None, weights = None) -> xr.DataArray:
+def vars_weighted_geomean(ds: xr.Dataset, vars=None, weights=None) -> xr.DataArray:
     if vars is None:
         vars = list(ds.data_vars)
     if weights is None:
         weights = np.ones(len(vars))
 
-    s = sum([np.log(ds[v]) * w for v, w in zip(vars, weights)])
-    da : xr.DataArray = np.exp(s / sum(weights))
-    return da.assign_attrs({
-        'method': 'Weighted Geometric Mean',
-        'description': f"Weighted Geometric Mean of variables: {', '.join([f'{v} ({w})' for v, w in zip(vars, weights)])}."
-    }).rename('weighted_geometric_mean')
+    s = sum([np.log(ds[v]) * w for v, w in zip(vars, weights, strict=False)])
+    da: xr.DataArray = np.exp(s / sum(weights))
+    return da.assign_attrs(
+        {
+            "method": "Weighted Geometric Mean",
+            "description": (
+                "Weighted Geometric Mean of variables: "
+                f"{', '.join([f'{v} ({w})' for v, w in zip(vars, weights, strict=False)])}."
+            )
+        }
+    ).rename("weighted_geometric_mean")
 
-def vars_geomean(ds: xr.Dataset, vars = None) -> xr.DataArray:
-    if vars is None: vars = list(ds.data_vars)
+
+def vars_geomean(ds: xr.Dataset, vars=None) -> xr.DataArray:
+    if vars is None:
+        vars = list(ds.data_vars)
     da = vars_weighted_geomean(ds, vars=vars)
-    return da.assign_attrs({'method': 'Geometric Mean', 'description': f"Geometric Mean of variables: {', '.join(vars)}."}).rename('geometric_mean')
+    return da.assign_attrs(
+        {"method": "Geometric Mean", "description": f"Geometric Mean of variables: {', '.join(vars)}."}
+    ).rename("geometric_mean")
+
 
-def limiting_factor(ds: xr.Dataset, vars = None, limiting_var: Optional[bool] = True) -> xr.DataArray | xr.Dataset:
+def limiting_factor(ds: xr.Dataset, vars=None, limiting_var: bool | None = True) -> xr.DataArray | xr.Dataset:
     if vars is None:
         vars = list(ds.data_vars)
-    
+
     da = ds[vars].to_array()
-    mask = da.notnull().all(dim='variable')
+    mask = da.notnull().all(dim="variable")
 
-    lim = da.min(dim='variable', skipna=True).where(mask).rename('limiting_factor')
-    lim = lim.assign_attrs({'method': 'Limiting Factor', 'description': f"Value of limiting factor among variables: {', '.join(vars)}."})
+    lim = da.min(dim="variable", skipna=True).where(mask).rename("limiting_factor")
+    lim = lim.assign_attrs(
+        {"method": "Limiting Factor", "description": f"Value of limiting factor among variables: {', '.join(vars)}."}
+    )
     if limiting_var:
-        lim_var = da.fillna(9999).argmin(dim='variable', skipna=True).where(mask).rename('limiting_var')
-        lim_var.attrs = {'method': 'Limiting Factor',
-                         'description': f"Limiting factor among: {', '.join(vars)}.",
-                         'legend': {f'{i}': v for i, v in enumerate(vars)}}
+        lim_var = da.fillna(9999).argmin(dim="variable", skipna=True).where(mask).rename("limiting_var")
+        lim_var.attrs = {
+            "method": "Limiting Factor",
+            "description": f"Limiting factor among: {', '.join(vars)}.",
+            "legend": {f"{i}": v for i, v in enumerate(vars)},
+        }
         return xr.merge([lim, lim_var])
     return lim.to_dataset()
 
-def _aggregate_vars(ds : xr.Dataset, method: str = 'mean', vars = None, weights = None, **kwargs) -> xr.DataArray | xr.Dataset:
-    if method.lower() == 'mean':
+
+def _aggregate_vars(
+    ds: xr.Dataset, method: str = "mean", vars=None, weights=None, **kwargs
+) -> xr.DataArray | xr.Dataset:
+    if method.lower() == "mean":
         return vars_mean(ds, vars=vars)
-    elif method.lower() == 'weighted_mean':
+    elif method.lower() == "weighted_mean":
         return vars_weighted_mean(ds, vars=vars, weights=weights)
-    elif method.lower() == 'geomean':
+    elif method.lower() == "geomean":
         return vars_geomean(ds, vars=vars)
-    elif method.lower() == 'weighted_geomean':
+    elif method.lower() == "weighted_geomean":
         return vars_weighted_geomean(ds, vars=vars, weights=weights)
-    elif method.lower() == 'limit_factor':
+    elif method.lower() == "limit_factor":
         return limiting_factor(ds, vars=vars, **kwargs)
     else:
         raise ValueError(f"Method '{method}' not recognized.")

lsapy/statistics.py

@@ -1,67 +1,70 @@
-from typing import Any, Optional, Union
+"""Statistics Module"""
+
+from typing import Any
+
+import geopandas as gpd
 import numpy as np
 import pandas as pd
-import geopandas as gpd
-import xarray as xr
 import regionmask
+import xarray as xr
 
 
 def statistics_summary(
         data: xr.DataArray | xr.Dataset,
-        on_vars: Optional[list] = None,
-        on_dims: Optional[list] = None,
-        on_dim_values: Optional[dict[str, Any]] = None,
-        bins : Optional[Union[list, np.ndarray]] = None,
-        bins_labels : Optional[list] = None,
-        all_bins : Optional[bool] = False,
-        cell_area : Optional[tuple[Union[float, str], str]] = None,
-        dropna: Optional[bool] = False,
+        on_vars: list | None = None,
+        on_dims: list | None = None,
+        on_dim_values: dict[str, Any] | None = None,
+        bins: list | np.ndarray | None = None,
+        bins_labels: list | None = None,
+        all_bins: bool | None = False,
+        cell_area: tuple[float | str, str] | None = None,
+        dropna: bool | None = False,
         **kwargs) -> pd.DataFrame:
-    
+
     def _correct_lowest_cut_interal(x: pd.Series) -> pd.Series:
-            first_cat = x.cat.categories[0]
-            lf = first_cat.left + (abs(first_cat.left) - bins[0])
-            return x.cat.rename_categories({first_cat: pd.Interval(lf, first_cat.right, closed='both')})
-    
-    
-    if bins is not None and bins_labels is not None and (len(bins)-1 != len(bins_labels)):
+        first_cat = x.cat.categories[0]
+        lf = first_cat.left + (abs(first_cat.left) - bins[0])
+        return x.cat.rename_categories({first_cat: pd.Interval(lf, first_cat.right, closed='both')})
+
+    if bins is not None and bins_labels is not None and (len(bins) - 1 != len(bins_labels)):
         raise ValueError('bins and bins_labels must have the same length')
-    
+
     if on_dim_values is not None:
         for dim, value in on_dim_values.items():
             data = data.sel({dim: value})
-    
+
     if on_vars is None:
         on_vars = list(data.data_vars)
     data = data[on_vars]
     if on_dims is None:
         on_dims = list(data.dims)
-        on_dims = [d for d in on_dims if d not in ['lat', 'lon', 'x', 'y']] # remove spatial dims
+        on_dims = [d for d in on_dims if d not in ['lat', 'lon', 'x', 'y']]  # remove spatial dims
     if cell_area:
         cell_area, cell_unit = cell_area
 
-
     df = data.to_dataframe().reset_index()
     if len(on_dims) > 0:
         df = df.drop(columns=[c for c in data.coords if c not in on_dims])
-    
+
     df = df.melt(id_vars=on_dims)
     _dims = ['variable'] + on_dims
 
     if bins is not None:
         df['bin'] = pd.cut(df['value'], bins=pd.Index(bins), **kwargs)
-        if 'include_lowest' in kwargs and kwargs['include_lowest'] == True:
+        if 'include_lowest' in kwargs and kwargs['include_lowest']:
             df['bin'] = _correct_lowest_cut_interal(df['bin'])
 
         if bins_labels is not None:
-            lab_mapping = dict(zip(df['bin'].cat.categories.astype(str), bins_labels))
+            lab_mapping = dict(zip(df['bin'].cat.categories.astype(str), bins_labels, strict=False))
         _dims.append('bin')
         if all_bins:
-            all_bins_inter = pd.Interval(df['bin'].cat.categories[0].left, df['bin'].cat.categories[-1].right, closed='both')
+            all_bins_inter = pd.Interval(
+                df['bin'].cat.categories[0].left, df['bin'].cat.categories[-1].right, closed='both'
+            )
             df_ = df.drop(columns=['bin']).assign(bin=all_bins_inter)
             df_.loc[df['value'].isnull(), 'bin'] = np.nan
             if bins_labels is not None:
-                lab_mapping.update({str(all_bins_inter) : 'all'})
+                lab_mapping.update({str(all_bins_inter): 'all'})
             df = pd.concat([df, df_])
         df['bin'] = df['bin'].astype(str)
 
@@ -73,7 +76,7 @@ def statistics_summary(
 
     if cell_area:
         df_out[f'area_{cell_unit}'] = df_out['count'] * cell_area
-    
+
     if dropna:
         return df_out.dropna()
     return df_out
@@ -81,19 +84,24 @@ def statistics_summary(
 
 def spatial_statistics_summary(
         data: xr.DataArray | xr.Dataset,
-        areas : gpd.GeoDataFrame,
-        name : Optional[str] = 'area',
-        on_vars: Optional[list] = None,
-        on_dims: Optional[list] = None,
-        on_dim_values: Optional[dict[str, Any]] = None,
-        bins : Optional[np.ndarray] = None,
-        bins_labels : Optional[list] = None,
-        all_bins : Optional[bool] = False,
-        cell_area : Optional[tuple[Union[float, str], str]] = None,
-        dropna: Optional[bool] = False,
-        mask_kwargs : dict = {},
-        stats_kwargs: dict = {}) -> pd.DataFrame:
-    
+        areas: gpd.GeoDataFrame,
+        name: str | None = 'area',
+        on_vars: list | None = None,
+        on_dims: list | None = None,
+        on_dim_values: dict[str, Any] | None = None,
+        bins: np.ndarray | None = None,
+        bins_labels: list | None = None,
+        all_bins: bool | None = False,
+        cell_area: tuple[float | str, str] | None = None,
+        dropna: bool | None = False,
+        mask_kwargs: dict = None,
+        stats_kwargs: dict = None) -> pd.DataFrame:
+
+    if mask_kwargs is None:
+        mask_kwargs = {}
+    if stats_kwargs is None:
+        stats_kwargs = {}
+
     regions = regionmask.from_geopandas(areas, name=name, **mask_kwargs)
     mask = regions.mask_3D(data)
 
@@ -102,8 +110,8 @@ def spatial_statistics_summary(
         df = statistics_summary(
             data.where(mask.sel(region=r)), on_vars=on_vars, on_dims=on_dims, on_dim_values=on_dim_values,
             bins=bins, bins_labels=bins_labels, all_bins=all_bins,
-            cell_area=cell_area ,dropna=dropna, **stats_kwargs
+            cell_area=cell_area, dropna=dropna, **stats_kwargs
         )
         df.insert(0, name, regions[r].name)
         out.append(df)
-    return pd.concat(out)
+    return pd.concat(out)

lsapy/utils.py

@@ -1,28 +1,24 @@
-"""
-Miscellaneous Utilities
-===============================
-
-...
-"""
+"""Miscellaneous Utilities"""
 
 from pathlib import Path
 
 import xarray as xr
 
+
 def _load_data(name: str) -> xr.Dataset:
     """Load a sample dataset for testing or example purposes."""
-    if name.lower() not in ['soil', 'climate']:
+    if name.lower() not in ["soil", "climate"]:
         raise ValueError(f"Invalid data name: {name}. Must be one of 'soil' or 'climate'.")
-    
-    data_path = Path(__file__).parent / 'data'
-    return xr.open_dataset(data_path / f'{name.lower()}_sample.nc')
+
+    data_path = Path(__file__).parent / "data"
+    return xr.open_dataset(data_path / f"{name.lower()}_sample.nc")
 
 
 def load_soil_data() -> xr.Dataset:
     """Load soil dataset for testing or example purposes."""
-    return _load_data('soil')
+    return _load_data("soil")
 
 
 def load_climate_data() -> xr.Dataset:
     """Load climate dataset for testing or example purposes."""
-    return _load_data('climate')
+    return _load_data("climate")

{lsapy-0.1.0.dev0.dist-info → lsapy-0.1.0.dev1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lsapy
-Version: 0.1.0.dev0
+Version: 0.1.0.dev1
 Summary: A Python package for Land Suitability Analysis.
 Keywords: LSAPy,Land Suitability Analysis,Spatial Analysis,GIS,Geography
 Author-email: Baptiste Hamon <baptiste.hamon@pg.canterbury.ac.nz>

lsapy-0.1.0.dev1.dist-info/RECORD

@@ -0,0 +1,12 @@
+lsapy/__init__.py,sha256=SreXtKw-kHRCBJB4s2JLW4IeJlgQzwpzKMZH4hcdzA8,286
+lsapy/criteria.py,sha256=7VFl1YQcZ015KELsLPsRFFEvGyvaj07Vkj79sM4t7p0,2952
+lsapy/functions.py,sha256=VaQS4ONrKWN5iAXT_xK_Ess1QXvodGsWzlBk8q-PO88,8513
+lsapy/lsa.py,sha256=AewMwvNzYdL3yGBAuncxy35G4oDybMaHqWGfWCxECNo,14459
+lsapy/statistics.py,sha256=lovmy94-km12ObcDj4TlrVpyZDLCvJTqyLHIgRWF0mw,4313
+lsapy/utils.py,sha256=y8-WL2wrPPIHeCwrqXERyY-X71qXnMAosUAmyUGjqU4,732
+lsapy/data/climate_sample.nc,sha256=ng2o7BxKLgzQWIl5fdiGWvdCQSeHeOO5-X3ciyD72Xo,52687793
+lsapy/data/soil_sample.nc,sha256=y7NpYV9VZKXwiYExynA0vF1KAn5rhsWqzwGJ4tFE-TQ,2535031
+lsapy-0.1.0.dev1.dist-info/licenses/LICENSE,sha256=6kbiFSfobTZ7beWiKnHpN902HgBx-Jzgcme0SvKqhKY,1091
+lsapy-0.1.0.dev1.dist-info/WHEEL,sha256=_2ozNFCLWc93bK4WKHCO-eDUENDlo-dgc9cU3qokYO4,82
+lsapy-0.1.0.dev1.dist-info/METADATA,sha256=kyssK7TWoRG-yNKHLbfW80y1wXmglG8emgF_KkV-dgQ,1145
+lsapy-0.1.0.dev1.dist-info/RECORD,,

lsapy-0.1.0.dev0.dist-info/RECORD

@@ -1,12 +0,0 @@
-lsapy/__init__.py,sha256=MXu8EIXhOrPg7Aba-9nrQCU51QTRAWNKxDapzYcnzFs,284
-lsapy/criteria.py,sha256=f2piZaEecKsLBvIH4JhLNXDNfNKNXUp70r23oeHErcQ,2971
-lsapy/functions.py,sha256=WYfpUREOsrkv-70wme_zL_HArM_AI2b5YxU8vtee5NI,8382
-lsapy/lsa.py,sha256=sBx96OpTDeaSrcX9s8IB2k91CNl0WUIW9hVM6oP-Hqg,14339
-lsapy/statistics.py,sha256=62O57pqF1Pth4s1D0MbQEyobnxDGZ2bNatC2uVFnHH8,4269
-lsapy/utils.py,sha256=eezLr5l0gqqgFEglzsSnbwx7JqOlVeKK3G63qllAJrg,776
-lsapy/data/climate_sample.nc,sha256=ng2o7BxKLgzQWIl5fdiGWvdCQSeHeOO5-X3ciyD72Xo,52687793
-lsapy/data/soil_sample.nc,sha256=y7NpYV9VZKXwiYExynA0vF1KAn5rhsWqzwGJ4tFE-TQ,2535031
-lsapy-0.1.0.dev0.dist-info/licenses/LICENSE,sha256=6kbiFSfobTZ7beWiKnHpN902HgBx-Jzgcme0SvKqhKY,1091
-lsapy-0.1.0.dev0.dist-info/WHEEL,sha256=_2ozNFCLWc93bK4WKHCO-eDUENDlo-dgc9cU3qokYO4,82
-lsapy-0.1.0.dev0.dist-info/METADATA,sha256=Vr0peYa--97JnnQKHtAsNb8SdreyYCs7espEr4XSemg,1145
-lsapy-0.1.0.dev0.dist-info/RECORD,,