stopped developing gessamans rule based multivariate LSx and instead …

…switched to decision tree
d3group · Aug 27, 2024 · 765184c · 765184c
1 parent 620d04b
commit 765184c
Show file tree

Hide file tree

Showing 13 changed files with 868 additions and 11 deletions.
diff --git a/_proc/02_levelSetKDEx_multivariate.ipynb b/_proc/02_levelSetKDEx_multivariate.ipynb
@@ -59,7 +59,7 @@
       "text/markdown": [
        "---\n",
        "\n",
-       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L31){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L33){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### LevelSetKDEx_multivariate\n",
        "\n",
@@ -83,7 +83,7 @@
       "text/plain": [
        "---\n",
        "\n",
-       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L31){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L33){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### LevelSetKDEx_multivariate\n",
        "\n",
@@ -135,7 +135,7 @@
       "text/markdown": [
        "---\n",
        "\n",
-       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L262){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L264){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### LevelSetKDEx_multivariate_opt\n",
        "\n",
@@ -171,7 +171,7 @@
       "text/plain": [
        "---\n",
        "\n",
-       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L262){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L264){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
        "\n",
        "### LevelSetKDEx_multivariate_opt\n",
        "\n",
@@ -216,6 +216,158 @@
     "show_doc(LevelSetKDEx_multivariate_opt)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "a831ecfe",
+   "metadata": {},
+   "source": [
+    "## Level-Set Approach based on Decision Tree"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "has_sd": true,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/markdown": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L492){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### LevelSetKDEx_DT\n",
+       "\n",
+       ">      LevelSetKDEx_DT (estimator, max_depth:int=8, min_samples_leaf:int=100)\n",
+       "\n",
+       "*[`LevelSetKDEx`](https://kaiguender.github.io/dddex/levelsetkdex_univariate.html#levelsetkdex) turns any point forecasting model into an estimator of the underlying conditional density.\n",
+       "The name 'LevelSet' stems from the fact that this approach interprets the values of the point forecasts\n",
+       "as a similarity measure between samples. \n",
+       "TBD*\n",
+       "\n",
+       "|    | **Type** | **Default** | **Details** |\n",
+       "| -- | -------- | ----------- | ----------- |\n",
+       "| estimator |  |  | Model with a .fit and .predict-method (implementing the scikit-learn estimator interface). |\n",
+       "| max_depth | int | 8 | Maximum depth of the decision tree used to generate the bins. |\n",
+       "| min_samples_leaf | int | 100 | Minimum number of samples required to be in a bin. |"
+      ],
+      "text/plain": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L492){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### LevelSetKDEx_DT\n",
+       "\n",
+       ">      LevelSetKDEx_DT (estimator, max_depth:int=8, min_samples_leaf:int=100)\n",
+       "\n",
+       "*`LevelSetKDEx` turns any point forecasting model into an estimator of the underlying conditional density.\n",
+       "The name 'LevelSet' stems from the fact that this approach interprets the values of the point forecasts\n",
+       "as a similarity measure between samples. \n",
+       "TBD*\n",
+       "\n",
+       "|    | **Type** | **Default** | **Details** |\n",
+       "| -- | -------- | ----------- | ----------- |\n",
+       "| estimator |  |  | Model with a .fit and .predict-method (implementing the scikit-learn estimator interface). |\n",
+       "| max_depth | int | 8 | Maximum depth of the decision tree used to generate the bins. |\n",
+       "| min_samples_leaf | int | 100 | Minimum number of samples required to be in a bin. |"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#| echo: false\n",
+    "#| output: asis\n",
+    "show_doc(LevelSetKDEx_DT)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "30884783",
+   "metadata": {},
+   "source": [
+    "## LSx Multivariate Gessaman Rule"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "has_sd": true,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/markdown": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L612){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### LevelSetKDEx_multivariate_bin\n",
+       "\n",
+       ">      LevelSetKDEx_multivariate_bin (estimator, nBinsPerDim:int=None)\n",
+       "\n",
+       "*[`LevelSetKDEx`](https://kaiguender.github.io/dddex/levelsetkdex_univariate.html#levelsetkdex) turns any point forecasting model into an estimator of the underlying conditional density.\n",
+       "The name 'LevelSet' stems from the fact that this approach interprets the values of the point forecasts\n",
+       "as a similarity measure between samples. \n",
+       "In this version of the LSx algorithm, we are applying the so-called Gessaman rule to create statistically\n",
+       "equivalent blocks of samples. In essence, the algorithm is a multivariate extension of the univariate\n",
+       "LevelSetKDEx algorithm based on bin-building. \n",
+       "We are creating equally sized bins of samples based on the point predictions of the samples specified via `X`\n",
+       "for every coordinate axis. Every bin of one axis is combined with the bins of all other axes resulting in\n",
+       "a total of nBinsPerDim^dim many bins. \n",
+       "Example: Let's say we have 100000 samples, the binSize is given as 20 and the number of dimension\n",
+       "is 3. As the binSize is given as 20, we want to create 5000 bins alltogether. Hence, there have to be\n",
+       "5000^(1/dim) = 5000^(1/3) = 17 bins per dimension. \n",
+       "IMPORTANT NOTE: The getWeights function is not yet finished and has to be completed.*\n",
+       "\n",
+       "|    | **Type** | **Default** | **Details** |\n",
+       "| -- | -------- | ----------- | ----------- |\n",
+       "| estimator |  |  | Model with a .fit and .predict-method (implementing the scikit-learn estimator interface). |\n",
+       "| nBinsPerDim | int | None | Number of samples belonging to each bin. |"
+      ],
+      "text/plain": [
+       "---\n",
+       "\n",
+       "[source](https://github.com/kaiguender/dddex/blob/main/dddex/levelSetKDEx_multivariate.py#L612){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+       "\n",
+       "### LevelSetKDEx_multivariate_bin\n",
+       "\n",
+       ">      LevelSetKDEx_multivariate_bin (estimator, nBinsPerDim:int=None)\n",
+       "\n",
+       "*`LevelSetKDEx` turns any point forecasting model into an estimator of the underlying conditional density.\n",
+       "The name 'LevelSet' stems from the fact that this approach interprets the values of the point forecasts\n",
+       "as a similarity measure between samples. \n",
+       "In this version of the LSx algorithm, we are applying the so-called Gessaman rule to create statistically\n",
+       "equivalent blocks of samples. In essence, the algorithm is a multivariate extension of the univariate\n",
+       "LevelSetKDEx algorithm based on bin-building. \n",
+       "We are creating equally sized bins of samples based on the point predictions of the samples specified via `X`\n",
+       "for every coordinate axis. Every bin of one axis is combined with the bins of all other axes resulting in\n",
+       "a total of nBinsPerDim^dim many bins. \n",
+       "Example: Let's say we have 100000 samples, the binSize is given as 20 and the number of dimension\n",
+       "is 3. As the binSize is given as 20, we want to create 5000 bins alltogether. Hence, there have to be\n",
+       "5000^(1/dim) = 5000^(1/3) = 17 bins per dimension. \n",
+       "IMPORTANT NOTE: The getWeights function is not yet finished and has to be completed.*\n",
+       "\n",
+       "|    | **Type** | **Default** | **Details** |\n",
+       "| -- | -------- | ----------- | ----------- |\n",
+       "| estimator |  |  | Model with a .fit and .predict-method (implementing the scikit-learn estimator interface). |\n",
+       "| nBinsPerDim | int | None | Number of samples belonging to each bin. |"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#| echo: false\n",
+    "#| output: asis\n",
+    "show_doc(LevelSetKDEx_multivariate_bin)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "c0d8dc43-8377-41b9-a5a7-aa5904038841",
@@ -295,6 +447,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "id": "5faa5708",
    "metadata": {
     "language": "python"
    },

diff --git a/dddex/__pycache__/__init__.cpython-39.pyc b/dddex/__pycache__/__init__.cpython-39.pyc
diff --git a/dddex/__pycache__/_modidx.cpython-39.pyc b/dddex/__pycache__/_modidx.cpython-39.pyc
diff --git a/dddex/__pycache__/baseClasses.cpython-39.pyc b/dddex/__pycache__/baseClasses.cpython-39.pyc
diff --git a/dddex/__pycache__/crossValidation.cpython-39.pyc b/dddex/__pycache__/crossValidation.cpython-39.pyc
diff --git a/dddex/__pycache__/levelSetKDEx_multivariate.cpython-39.pyc b/dddex/__pycache__/levelSetKDEx_multivariate.cpython-39.pyc
diff --git a/dddex/__pycache__/levelSetKDEx_univariate.cpython-39.pyc b/dddex/__pycache__/levelSetKDEx_univariate.cpython-39.pyc
diff --git a/dddex/__pycache__/loadData.cpython-39.pyc b/dddex/__pycache__/loadData.cpython-39.pyc
diff --git a/dddex/__pycache__/utils.cpython-39.pyc b/dddex/__pycache__/utils.cpython-39.pyc
diff --git a/dddex/__pycache__/wSAA.cpython-39.pyc b/dddex/__pycache__/wSAA.cpython-39.pyc
diff --git a/dddex/_modidx.py b/dddex/_modidx.py
@@ -61,7 +61,15 @@
                                                                                           'dddex/crossValidation.py'),
                                        'dddex.crossValidation.groupedTimeSeriesSplit': ( 'crossvalidation.html#groupedtimeseriessplit',
                                                                                          'dddex/crossValidation.py')},
-            'dddex.levelSetKDEx_multivariate': { 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate',
+            'dddex.levelSetKDEx_multivariate': { 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_DT': ( 'levelsetkdex_multivariate.html#levelsetkdex_dt',
+                                                                                                      'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_DT.__init__': ( 'levelsetkdex_multivariate.html#levelsetkdex_dt.__init__',
+                                                                                                               'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_DT.fit': ( 'levelsetkdex_multivariate.html#levelsetkdex_dt.fit',
+                                                                                                          'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_DT.getWeights': ( 'levelsetkdex_multivariate.html#levelsetkdex_dt.getweights',
+                                                                                                                 'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate',
                                                                                                                 'dddex/levelSetKDEx_multivariate.py'),
                                                  'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate.__init__': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate.__init__',
                                                                                                                          'dddex/levelSetKDEx_multivariate.py'),
@@ -71,6 +79,14 @@
                                                                                                                     'dddex/levelSetKDEx_multivariate.py'),
                                                  'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate.getWeights': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate.getweights',
                                                                                                                            'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_bin': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_bin',
+                                                                                                                    'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_bin.__init__': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_bin.__init__',
+                                                                                                                             'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_bin.fit': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_bin.fit',
+                                                                                                                        'dddex/levelSetKDEx_multivariate.py'),
+                                                 'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_bin.getWeights': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_bin.getweights',
+                                                                                                                               'dddex/levelSetKDEx_multivariate.py'),
                                                  'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_opt': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_opt',
                                                                                                                     'dddex/levelSetKDEx_multivariate.py'),
                                                  'dddex.levelSetKDEx_multivariate.LevelSetKDEx_multivariate_opt.__init__': ( 'levelsetkdex_multivariate.html#levelsetkdex_multivariate_opt.__init__',