py-geodetector 0.1.2__tar.gz → 0.1.3__tar.gz

{py_geodetector-0.1.2 → py_geodetector-0.1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: py_geodetector
-Version: 0.1.2
+Version: 0.1.3
 Summary: A simple Python package for the geodetector
 Project-URL: Homepage, https://github.com/djw-easy/GeoDetector
 Project-URL: Bug Tracker, https://github.com/djw-easy/GeoDetector/issues
@@ -41,6 +41,9 @@ interaction_df, interaction_relationship_df = gd.interaction_detector(relationsh
 # ecological detect
 ecological_df = gd.ecological_detector()
 
+# risk detect
+risk_result = gd.risk_detector()
+
 # plot 
 # use a heatmap visualize the interaction detect result, 
 # red text means that the ecological detection results show a significant difference

{py_geodetector-0.1.2 → py_geodetector-0.1.3}/README.md

@@ -28,6 +28,9 @@ interaction_df, interaction_relationship_df = gd.interaction_detector(relationsh
 # ecological detect
 ecological_df = gd.ecological_detector()
 
+# risk detect
+risk_result = gd.risk_detector()
+
 # plot 
 # use a heatmap visualize the interaction detect result, 
 # red text means that the ecological detection results show a significant difference

{py_geodetector-0.1.2 → py_geodetector-0.1.3}/example.ipynb

@@ -13,6 +13,15 @@
    "cell_type": "code",
    "execution_count": 2,
    "metadata": {},
+   "outputs": [],
+   "source": [
+    "from src.py_geodetector import GeoDetector, load_example_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -90,7 +99,7 @@
        "4       6.49     3       2      4"
       ]
      },
-     "execution_count": 2,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -102,7 +111,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -111,7 +120,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -163,7 +172,7 @@
        "p value      0.363236  0.000117  0.040804"
       ]
      },
-     "execution_count": 4,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -174,7 +183,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -233,7 +242,7 @@
        "level   0.663524  0.713597  0.606709"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -244,7 +253,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -303,7 +312,7 @@
        "level   Enhance, bi-  Enhance, bi-   NaN"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -314,7 +323,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -373,7 +382,7 @@
        "level     Y      N    NaN"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -384,7 +393,138 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "r = gd.risk_detector()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "type\n",
+       "1    6.340000\n",
+       "2    6.687500\n",
+       "3    6.583279\n",
+       "5    5.843810\n",
+       "7    6.347073\n",
+       "Name: incidence, dtype: float64"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "r['type']['risk']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>1</th>\n",
+       "      <th>2</th>\n",
+       "      <th>3</th>\n",
+       "      <th>5</th>\n",
+       "      <th>7</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>True</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>True</td>\n",
+       "      <td>False</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>False</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>True</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       1      2     3     5   7\n",
+       "1    NaN    NaN   NaN   NaN NaN\n",
+       "2   True    NaN   NaN   NaN NaN\n",
+       "3   True  False   NaN   NaN NaN\n",
+       "5   True   True  True   NaN NaN\n",
+       "7  False   True  True  True NaN"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "r['type']['ttest_stra']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {

{py_geodetector-0.1.2 → py_geodetector-0.1.3}/pyproject.toml

@@ -11,7 +11,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "py_geodetector"
-version = "0.1.2"
+version = "0.1.3"
 authors = [
     { name = "djw", email = "djweasy@163.com" },
 ]

{py_geodetector-0.1.2 → py_geodetector-0.1.3}/src/py_geodetector/geodetector.py

@@ -2,8 +2,8 @@ import warnings
 import numpy as np
 import pandas as pd
 from typing import Sequence
-from scipy.stats import f, ncf
 import matplotlib.pyplot as plt
+from scipy.stats import f, levene, ncf, ttest_ind
 
 
 from pathlib import Path
@@ -27,10 +27,11 @@ def _plot_value(ax, interaction_df, ecological_df, value_fontsize=10):
 
 
 class GeoDetector(object):
-    def __init__(self, df: pd.DataFrame, y: str, factors: Sequence[str]):
+    def __init__(self, df: pd.DataFrame, y: str, factors: Sequence[str], alpha=0.05):
         self.df = df
         self.y = y
         self.factors = factors
+        self.alpha = alpha
         self._check_data(df, y, factors)
         self.factor_df, self.interaction_df, self.ecological_df = None, None, None
 
@@ -155,11 +156,44 @@ class GeoDetector(object):
                 ssw2, _, _ = self._cal_ssw(self.df, self.y, self.factors[j])
                 dfd = self.df[self.factors[j]].notna().sum()-1
                 fval = (dfn*(dfd-1)*ssw1)/(dfd*(dfn-1)*ssw2)
-                if fval<f.ppf(0.05, dfn, dfn):
+                if fval<f.ppf(self.alpha, dfn, dfn):
                     self.ecological_df.loc[self.factors[i], self.factors[j]] = 'Y'
                 else:
                     self.ecological_df.loc[self.factors[i], self.factors[j]] = 'N'
         return self.ecological_df
+    
+    def risk_detector(self):
+        """
+        Compares the difference of average values between sub-groups
+        Reference:
+            https://github.com/gsnrguo/QGIS-Geographical-detector/blob/main/gd_core/geodetector.py
+        """
+        risk_result = dict()
+        for factor in self.factors:
+            risk_name = self.df.groupby(factor)[self.y].mean()
+            strata = np.sort(self.df[factor].unique())
+            t_test = np.empty((len(strata), len(strata)))
+            t_test.fill(np.nan)
+            t_test_strata = pd.DataFrame(t_test, index=strata, columns=strata)
+            for i in range(len(strata) - 1):
+                for j in range(i + 1, len(strata)):
+                    y_i = self.df.loc[self.df[factor] == strata[i], [self.y]]
+                    y_j = self.df.loc[self.df[factor] == strata[j], [self.y]]
+                    y_i = np.array(y_i).reshape(-1)
+                    y_j = np.array(y_j).reshape(-1)
+                    # hypothesis testing of variance homogeneity
+                    levene_result = levene(y_i, y_j)
+                    if levene_result.pvalue < self.alpha:
+                        # variance non-homogeneous
+                        ttest_result = ttest_ind(y_i, y_j, equal_var=False)
+                    else:
+                        ttest_result = ttest_ind(y_i, y_j)
+
+                    t_test_strata.iloc[j, i] = ttest_result.pvalue <= self.alpha
+
+            risk_factor = dict(risk=risk_name, ttest_stra=t_test_strata)
+            risk_result[factor] = risk_factor
+        return risk_result
 
     def plot(self, tick_fontsize=10, value_fontsize=10, colorbar_fontsize=10, show=True):
         if isinstance(self.interaction_df, type(None)):
@@ -188,4 +222,4 @@ class GeoDetector(object):
             plt.show()
             return ax
         else:
-            return ax
+            return ax