You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
1143 lines
594 KiB
1143 lines
594 KiB
2 years ago
|
{
|
||
|
|
"cells": [
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"___\n",
|
||
|
|
"\n",
|
||
|
|
"<a href='https://www.udemy.com/user/joseportilla/'><img src='../Pierian_Data_Logo.png'/></a>\n",
|
||
|
|
"___\n",
|
||
|
|
"<center><em>Copyright by Pierian Data Inc.</em></center>\n",
|
||
|
|
"<center><em>For more information, visit us at <a href='http://www.pieriandata.com'>www.pieriandata.com</a></em></center>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"# Exploring Support Vector Machines\n",
|
||
|
|
"\n",
|
||
|
|
"**NOTE: For this example, we will explore the algorithm, so we'll skip scaling and also skip a train\\test split and instead see how the various parameters can change an SVM (easiest to visualize the effects in classification)**\n",
|
||
|
|
"\n",
|
||
|
|
"[Link to a great Paper on SVM](http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=4448154647BC7B10C991CEF2236BBA38?doi=10.1.1.114.4288&rep=rep1&type=pdf)\n",
|
||
|
|
"* A tutorial on support vector regression by ALEX J. SMOLA and BERNHARD SCHOLKOPF\n",
|
||
|
|
"\n",
|
||
|
|
"## SVM - Classification"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 1,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"import numpy as np\n",
|
||
|
|
"import pandas as pd\n",
|
||
|
|
"import seaborn as sns\n",
|
||
|
|
"import matplotlib.pyplot as plt"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## Data\n",
|
||
|
|
"\n",
|
||
|
|
"The data shown here simulates a medical study in which mice infected with a virus were given various doses of two medicines and then checked 2 weeks later to see if they were still infected. Given this data, our goal is to create a classifcation model than predict (given two dosage measurements) if they mouse will still be infected with the virus.\n",
|
||
|
|
"\n",
|
||
|
|
"You will notice the groups are very separable, this is on purpose, to explore how the various parameters of an SVM model behave."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 2,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"df = pd.read_csv(\"../DATA/mouse_viral_study.csv\")"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 3,
|
||
|
|
"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>Med_1_mL</th>\n",
|
||
|
|
" <th>Med_2_mL</th>\n",
|
||
|
|
" <th>Virus Present</th>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" </thead>\n",
|
||
|
|
" <tbody>\n",
|
||
|
|
" <tr>\n",
|
||
|
|
" <th>0</th>\n",
|
||
|
|
" <td>6.508231</td>\n",
|
||
|
|
" <td>8.582531</td>\n",
|
||
|
|
" <td>0</td>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" <tr>\n",
|
||
|
|
" <th>1</th>\n",
|
||
|
|
" <td>4.126116</td>\n",
|
||
|
|
" <td>3.073459</td>\n",
|
||
|
|
" <td>1</td>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" <tr>\n",
|
||
|
|
" <th>2</th>\n",
|
||
|
|
" <td>6.427870</td>\n",
|
||
|
|
" <td>6.369758</td>\n",
|
||
|
|
" <td>0</td>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" <tr>\n",
|
||
|
|
" <th>3</th>\n",
|
||
|
|
" <td>3.672953</td>\n",
|
||
|
|
" <td>4.905215</td>\n",
|
||
|
|
" <td>1</td>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" <tr>\n",
|
||
|
|
" <th>4</th>\n",
|
||
|
|
" <td>1.580321</td>\n",
|
||
|
|
" <td>2.440562</td>\n",
|
||
|
|
" <td>1</td>\n",
|
||
|
|
" </tr>\n",
|
||
|
|
" </tbody>\n",
|
||
|
|
"</table>\n",
|
||
|
|
"</div>"
|
||
|
|
],
|
||
|
|
"text/plain": [
|
||
|
|
" Med_1_mL Med_2_mL Virus Present\n",
|
||
|
|
"0 6.508231 8.582531 0\n",
|
||
|
|
"1 4.126116 3.073459 1\n",
|
||
|
|
"2 6.427870 6.369758 0\n",
|
||
|
|
"3 3.672953 4.905215 1\n",
|
||
|
|
"4 1.580321 2.440562 1"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 3,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"df.head()"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 4,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"Index(['Med_1_mL', 'Med_2_mL', 'Virus Present'], dtype='object')"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 4,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"df.columns"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## Classes"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 5,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"<AxesSubplot:xlabel='Med_1_mL', ylabel='Med_2_mL'>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 5,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"sns.scatterplot(x='Med_1_mL',y='Med_2_mL',hue='Virus Present',\n",
|
||
|
|
" data=df,palette='seismic')"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## Separating Hyperplane\n",
|
||
|
|
"\n",
|
||
|
|
"Our goal with SVM is to create the best separating hyperplane. In 2 dimensions, this is simply a line."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 6,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"[<matplotlib.lines.Line2D at 0x1dab706f208>]"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 6,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"sns.scatterplot(x='Med_1_mL',y='Med_2_mL',hue='Virus Present',palette='seismic',data=df)\n",
|
||
|
|
"\n",
|
||
|
|
"# We want to somehow automatically create a separating hyperplane ( a line in 2D)\n",
|
||
|
|
"\n",
|
||
|
|
"x = np.linspace(0,10,100)\n",
|
||
|
|
"m = -1\n",
|
||
|
|
"b = 11\n",
|
||
|
|
"y = m*x + b\n",
|
||
|
|
"plt.plot(x,y,'k')"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## SVM - Support Vector Machine"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 7,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"from sklearn.svm import SVC # Supprt Vector Classifier"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 8,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"name": "stdout",
|
||
|
|
"output_type": "stream",
|
||
|
|
"text": [
|
||
|
|
"Help on class SVC in module sklearn.svm._classes:\n",
|
||
|
|
"\n",
|
||
|
|
"class SVC(sklearn.svm._base.BaseSVC)\n",
|
||
|
|
" | SVC(*, C=1.0, kernel='rbf', degree=3, gamma='scale', coef0=0.0, shrinking=True, probability=False, tol=0.001, cache_size=200, class_weight=None, verbose=False, max_iter=-1, decision_function_shape='ovr', break_ties=False, random_state=None)\n",
|
||
|
|
" | \n",
|
||
|
|
" | C-Support Vector Classification.\n",
|
||
|
|
" | \n",
|
||
|
|
" | The implementation is based on libsvm. The fit time scales at least\n",
|
||
|
|
" | quadratically with the number of samples and may be impractical\n",
|
||
|
|
" | beyond tens of thousands of samples. For large datasets\n",
|
||
|
|
" | consider using :class:`sklearn.svm.LinearSVC` or\n",
|
||
|
|
" | :class:`sklearn.linear_model.SGDClassifier` instead, possibly after a\n",
|
||
|
|
" | :class:`sklearn.kernel_approximation.Nystroem` transformer.\n",
|
||
|
|
" | \n",
|
||
|
|
" | The multiclass support is handled according to a one-vs-one scheme.\n",
|
||
|
|
" | \n",
|
||
|
|
" | For details on the precise mathematical formulation of the provided\n",
|
||
|
|
" | kernel functions and how `gamma`, `coef0` and `degree` affect each\n",
|
||
|
|
" | other, see the corresponding section in the narrative documentation:\n",
|
||
|
|
" | :ref:`svm_kernels`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Read more in the :ref:`User Guide <svm_classification>`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | C : float, default=1.0\n",
|
||
|
|
" | Regularization parameter. The strength of the regularization is\n",
|
||
|
|
" | inversely proportional to C. Must be strictly positive. The penalty\n",
|
||
|
|
" | is a squared l2 penalty.\n",
|
||
|
|
" | \n",
|
||
|
|
" | kernel : {'linear', 'poly', 'rbf', 'sigmoid', 'precomputed'}, default='rbf'\n",
|
||
|
|
" | Specifies the kernel type to be used in the algorithm.\n",
|
||
|
|
" | It must be one of 'linear', 'poly', 'rbf', 'sigmoid', 'precomputed' or\n",
|
||
|
|
" | a callable.\n",
|
||
|
|
" | If none is given, 'rbf' will be used. If a callable is given it is\n",
|
||
|
|
" | used to pre-compute the kernel matrix from data matrices; that matrix\n",
|
||
|
|
" | should be an array of shape ``(n_samples, n_samples)``.\n",
|
||
|
|
" | \n",
|
||
|
|
" | degree : int, default=3\n",
|
||
|
|
" | Degree of the polynomial kernel function ('poly').\n",
|
||
|
|
" | Ignored by all other kernels.\n",
|
||
|
|
" | \n",
|
||
|
|
" | gamma : {'scale', 'auto'} or float, default='scale'\n",
|
||
|
|
" | Kernel coefficient for 'rbf', 'poly' and 'sigmoid'.\n",
|
||
|
|
" | \n",
|
||
|
|
" | - if ``gamma='scale'`` (default) is passed then it uses\n",
|
||
|
|
" | 1 / (n_features * X.var()) as value of gamma,\n",
|
||
|
|
" | - if 'auto', uses 1 / n_features.\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. versionchanged:: 0.22\n",
|
||
|
|
" | The default value of ``gamma`` changed from 'auto' to 'scale'.\n",
|
||
|
|
" | \n",
|
||
|
|
" | coef0 : float, default=0.0\n",
|
||
|
|
" | Independent term in kernel function.\n",
|
||
|
|
" | It is only significant in 'poly' and 'sigmoid'.\n",
|
||
|
|
" | \n",
|
||
|
|
" | shrinking : bool, default=True\n",
|
||
|
|
" | Whether to use the shrinking heuristic.\n",
|
||
|
|
" | See the :ref:`User Guide <shrinking_svm>`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | probability : bool, default=False\n",
|
||
|
|
" | Whether to enable probability estimates. This must be enabled prior\n",
|
||
|
|
" | to calling `fit`, will slow down that method as it internally uses\n",
|
||
|
|
" | 5-fold cross-validation, and `predict_proba` may be inconsistent with\n",
|
||
|
|
" | `predict`. Read more in the :ref:`User Guide <scores_probabilities>`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | tol : float, default=1e-3\n",
|
||
|
|
" | Tolerance for stopping criterion.\n",
|
||
|
|
" | \n",
|
||
|
|
" | cache_size : float, default=200\n",
|
||
|
|
" | Specify the size of the kernel cache (in MB).\n",
|
||
|
|
" | \n",
|
||
|
|
" | class_weight : dict or 'balanced', default=None\n",
|
||
|
|
" | Set the parameter C of class i to class_weight[i]*C for\n",
|
||
|
|
" | SVC. If not given, all classes are supposed to have\n",
|
||
|
|
" | weight one.\n",
|
||
|
|
" | The \"balanced\" mode uses the values of y to automatically adjust\n",
|
||
|
|
" | weights inversely proportional to class frequencies in the input data\n",
|
||
|
|
" | as ``n_samples / (n_classes * np.bincount(y))``\n",
|
||
|
|
" | \n",
|
||
|
|
" | verbose : bool, default=False\n",
|
||
|
|
" | Enable verbose output. Note that this setting takes advantage of a\n",
|
||
|
|
" | per-process runtime setting in libsvm that, if enabled, may not work\n",
|
||
|
|
" | properly in a multithreaded context.\n",
|
||
|
|
" | \n",
|
||
|
|
" | max_iter : int, default=-1\n",
|
||
|
|
" | Hard limit on iterations within solver, or -1 for no limit.\n",
|
||
|
|
" | \n",
|
||
|
|
" | decision_function_shape : {'ovo', 'ovr'}, default='ovr'\n",
|
||
|
|
" | Whether to return a one-vs-rest ('ovr') decision function of shape\n",
|
||
|
|
" | (n_samples, n_classes) as all other classifiers, or the original\n",
|
||
|
|
" | one-vs-one ('ovo') decision function of libsvm which has shape\n",
|
||
|
|
" | (n_samples, n_classes * (n_classes - 1) / 2). However, one-vs-one\n",
|
||
|
|
" | ('ovo') is always used as multi-class strategy. The parameter is\n",
|
||
|
|
" | ignored for binary classification.\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. versionchanged:: 0.19\n",
|
||
|
|
" | decision_function_shape is 'ovr' by default.\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. versionadded:: 0.17\n",
|
||
|
|
" | *decision_function_shape='ovr'* is recommended.\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. versionchanged:: 0.17\n",
|
||
|
|
" | Deprecated *decision_function_shape='ovo' and None*.\n",
|
||
|
|
" | \n",
|
||
|
|
" | break_ties : bool, default=False\n",
|
||
|
|
" | If true, ``decision_function_shape='ovr'``, and number of classes > 2,\n",
|
||
|
|
" | :term:`predict` will break ties according to the confidence values of\n",
|
||
|
|
" | :term:`decision_function`; otherwise the first class among the tied\n",
|
||
|
|
" | classes is returned. Please note that breaking ties comes at a\n",
|
||
|
|
" | relatively high computational cost compared to a simple predict.\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. versionadded:: 0.22\n",
|
||
|
|
" | \n",
|
||
|
|
" | random_state : int or RandomState instance, default=None\n",
|
||
|
|
" | Controls the pseudo random number generation for shuffling the data for\n",
|
||
|
|
" | probability estimates. Ignored when `probability` is False.\n",
|
||
|
|
" | Pass an int for reproducible output across multiple function calls.\n",
|
||
|
|
" | See :term:`Glossary <random_state>`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Attributes\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | support_ : ndarray of shape (n_SV,)\n",
|
||
|
|
" | Indices of support vectors.\n",
|
||
|
|
" | \n",
|
||
|
|
" | support_vectors_ : ndarray of shape (n_SV, n_features)\n",
|
||
|
|
" | Support vectors.\n",
|
||
|
|
" | \n",
|
||
|
|
" | n_support_ : ndarray of shape (n_class,), dtype=int32\n",
|
||
|
|
" | Number of support vectors for each class.\n",
|
||
|
|
" | \n",
|
||
|
|
" | dual_coef_ : ndarray of shape (n_class-1, n_SV)\n",
|
||
|
|
" | Dual coefficients of the support vector in the decision\n",
|
||
|
|
" | function (see :ref:`sgd_mathematical_formulation`), multiplied by\n",
|
||
|
|
" | their targets.\n",
|
||
|
|
" | For multiclass, coefficient for all 1-vs-1 classifiers.\n",
|
||
|
|
" | The layout of the coefficients in the multiclass case is somewhat\n",
|
||
|
|
" | non-trivial. See the :ref:`multi-class section of the User Guide\n",
|
||
|
|
" | <svm_multi_class>` for details.\n",
|
||
|
|
" | \n",
|
||
|
|
" | coef_ : ndarray of shape (n_class * (n_class-1) / 2, n_features)\n",
|
||
|
|
" | Weights assigned to the features (coefficients in the primal\n",
|
||
|
|
" | problem). This is only available in the case of a linear kernel.\n",
|
||
|
|
" | \n",
|
||
|
|
" | `coef_` is a readonly property derived from `dual_coef_` and\n",
|
||
|
|
" | `support_vectors_`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | intercept_ : ndarray of shape (n_class * (n_class-1) / 2,)\n",
|
||
|
|
" | Constants in decision function.\n",
|
||
|
|
" | \n",
|
||
|
|
" | fit_status_ : int\n",
|
||
|
|
" | 0 if correctly fitted, 1 otherwise (will raise warning)\n",
|
||
|
|
" | \n",
|
||
|
|
" | classes_ : ndarray of shape (n_classes,)\n",
|
||
|
|
" | The classes labels.\n",
|
||
|
|
" | \n",
|
||
|
|
" | probA_ : ndarray of shape (n_class * (n_class-1) / 2)\n",
|
||
|
|
" | probB_ : ndarray of shape (n_class * (n_class-1) / 2)\n",
|
||
|
|
" | If `probability=True`, it corresponds to the parameters learned in\n",
|
||
|
|
" | Platt scaling to produce probability estimates from decision values.\n",
|
||
|
|
" | If `probability=False`, it's an empty array. Platt scaling uses the\n",
|
||
|
|
" | logistic function\n",
|
||
|
|
" | ``1 / (1 + exp(decision_value * probA_ + probB_))``\n",
|
||
|
|
" | where ``probA_`` and ``probB_`` are learned from the dataset [2]_. For\n",
|
||
|
|
" | more information on the multiclass case and training procedure see\n",
|
||
|
|
" | section 8 of [1]_.\n",
|
||
|
|
" | \n",
|
||
|
|
" | class_weight_ : ndarray of shape (n_class,)\n",
|
||
|
|
" | Multipliers of parameter C for each class.\n",
|
||
|
|
" | Computed based on the ``class_weight`` parameter.\n",
|
||
|
|
" | \n",
|
||
|
|
" | shape_fit_ : tuple of int of shape (n_dimensions_of_X,)\n",
|
||
|
|
" | Array dimensions of training vector ``X``.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Examples\n",
|
||
|
|
" | --------\n",
|
||
|
|
" | >>> import numpy as np\n",
|
||
|
|
" | >>> from sklearn.pipeline import make_pipeline\n",
|
||
|
|
" | >>> from sklearn.preprocessing import StandardScaler\n",
|
||
|
|
" | >>> X = np.array([[-1, -1], [-2, -1], [1, 1], [2, 1]])\n",
|
||
|
|
" | >>> y = np.array([1, 1, 2, 2])\n",
|
||
|
|
" | >>> from sklearn.svm import SVC\n",
|
||
|
|
" | >>> clf = make_pipeline(StandardScaler(), SVC(gamma='auto'))\n",
|
||
|
|
" | >>> clf.fit(X, y)\n",
|
||
|
|
" | Pipeline(steps=[('standardscaler', StandardScaler()),\n",
|
||
|
|
" | ('svc', SVC(gamma='auto'))])\n",
|
||
|
|
" | \n",
|
||
|
|
" | >>> print(clf.predict([[-0.8, -1]]))\n",
|
||
|
|
" | [1]\n",
|
||
|
|
" | \n",
|
||
|
|
" | See also\n",
|
||
|
|
" | --------\n",
|
||
|
|
" | SVR\n",
|
||
|
|
" | Support Vector Machine for Regression implemented using libsvm.\n",
|
||
|
|
" | \n",
|
||
|
|
" | LinearSVC\n",
|
||
|
|
" | Scalable Linear Support Vector Machine for classification\n",
|
||
|
|
" | implemented using liblinear. Check the See also section of\n",
|
||
|
|
" | LinearSVC for more comparison element.\n",
|
||
|
|
" | \n",
|
||
|
|
" | References\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | .. [1] `LIBSVM: A Library for Support Vector Machines\n",
|
||
|
|
" | <http://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.pdf>`_\n",
|
||
|
|
" | \n",
|
||
|
|
" | .. [2] `Platt, John (1999). \"Probabilistic outputs for support vector\n",
|
||
|
|
" | machines and comparison to regularizedlikelihood methods.\"\n",
|
||
|
|
" | <http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.41.1639>`_\n",
|
||
|
|
" | \n",
|
||
|
|
" | Method resolution order:\n",
|
||
|
|
" | SVC\n",
|
||
|
|
" | sklearn.svm._base.BaseSVC\n",
|
||
|
|
" | sklearn.base.ClassifierMixin\n",
|
||
|
|
" | sklearn.svm._base.BaseLibSVM\n",
|
||
|
|
" | sklearn.base.BaseEstimator\n",
|
||
|
|
" | builtins.object\n",
|
||
|
|
" | \n",
|
||
|
|
" | Methods defined here:\n",
|
||
|
|
" | \n",
|
||
|
|
" | __init__(self, *, C=1.0, kernel='rbf', degree=3, gamma='scale', coef0=0.0, shrinking=True, probability=False, tol=0.001, cache_size=200, class_weight=None, verbose=False, max_iter=-1, decision_function_shape='ovr', break_ties=False, random_state=None)\n",
|
||
|
|
" | Initialize self. See help(type(self)) for accurate signature.\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Data and other attributes defined here:\n",
|
||
|
|
" | \n",
|
||
|
|
" | __abstractmethods__ = frozenset()\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Methods inherited from sklearn.svm._base.BaseSVC:\n",
|
||
|
|
" | \n",
|
||
|
|
" | decision_function(self, X)\n",
|
||
|
|
" | Evaluates the decision function for the samples in X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : array-like of shape (n_samples, n_features)\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | X : ndarray of shape (n_samples, n_classes * (n_classes-1) / 2)\n",
|
||
|
|
" | Returns the decision function of the sample for each class\n",
|
||
|
|
" | in the model.\n",
|
||
|
|
" | If decision_function_shape='ovr', the shape is (n_samples,\n",
|
||
|
|
" | n_classes).\n",
|
||
|
|
" | \n",
|
||
|
|
" | Notes\n",
|
||
|
|
" | -----\n",
|
||
|
|
" | If decision_function_shape='ovo', the function values are proportional\n",
|
||
|
|
" | to the distance of the samples X to the separating hyperplane. If the\n",
|
||
|
|
" | exact distances are required, divide the function values by the norm of\n",
|
||
|
|
" | the weight vector (``coef_``). See also `this question\n",
|
||
|
|
" | <https://stats.stackexchange.com/questions/14876/\n",
|
||
|
|
" | interpreting-distance-from-hyperplane-in-svm>`_ for further details.\n",
|
||
|
|
" | If decision_function_shape='ovr', the decision function is a monotonic\n",
|
||
|
|
" | transformation of ovo decision function.\n",
|
||
|
|
" | \n",
|
||
|
|
" | predict(self, X)\n",
|
||
|
|
" | Perform classification on samples in X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | For an one-class model, +1 or -1 is returned.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : {array-like, sparse matrix} of shape (n_samples, n_features) or (n_samples_test, n_samples_train)\n",
|
||
|
|
" | For kernel=\"precomputed\", the expected shape of X is\n",
|
||
|
|
" | (n_samples_test, n_samples_train).\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | y_pred : ndarray of shape (n_samples,)\n",
|
||
|
|
" | Class labels for samples in X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Data descriptors inherited from sklearn.svm._base.BaseSVC:\n",
|
||
|
|
" | \n",
|
||
|
|
" | predict_log_proba\n",
|
||
|
|
" | Compute log probabilities of possible outcomes for samples in X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | The model need to have probability information computed at training\n",
|
||
|
|
" | time: fit with attribute `probability` set to True.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : array-like of shape (n_samples, n_features) or (n_samples_test, n_samples_train)\n",
|
||
|
|
" | For kernel=\"precomputed\", the expected shape of X is\n",
|
||
|
|
" | (n_samples_test, n_samples_train).\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | T : ndarray of shape (n_samples, n_classes)\n",
|
||
|
|
" | Returns the log-probabilities of the sample for each class in\n",
|
||
|
|
" | the model. The columns correspond to the classes in sorted\n",
|
||
|
|
" | order, as they appear in the attribute :term:`classes_`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Notes\n",
|
||
|
|
" | -----\n",
|
||
|
|
" | The probability model is created using cross validation, so\n",
|
||
|
|
" | the results can be slightly different than those obtained by\n",
|
||
|
|
" | predict. Also, it will produce meaningless results on very small\n",
|
||
|
|
" | datasets.\n",
|
||
|
|
" | \n",
|
||
|
|
" | predict_proba\n",
|
||
|
|
" | Compute probabilities of possible outcomes for samples in X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | The model need to have probability information computed at training\n",
|
||
|
|
" | time: fit with attribute `probability` set to True.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : array-like of shape (n_samples, n_features)\n",
|
||
|
|
" | For kernel=\"precomputed\", the expected shape of X is\n",
|
||
|
|
" | [n_samples_test, n_samples_train]\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | T : ndarray of shape (n_samples, n_classes)\n",
|
||
|
|
" | Returns the probability of the sample for each class in\n",
|
||
|
|
" | the model. The columns correspond to the classes in sorted\n",
|
||
|
|
" | order, as they appear in the attribute :term:`classes_`.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Notes\n",
|
||
|
|
" | -----\n",
|
||
|
|
" | The probability model is created using cross validation, so\n",
|
||
|
|
" | the results can be slightly different than those obtained by\n",
|
||
|
|
" | predict. Also, it will produce meaningless results on very small\n",
|
||
|
|
" | datasets.\n",
|
||
|
|
" | \n",
|
||
|
|
" | probA_\n",
|
||
|
|
" | \n",
|
||
|
|
" | probB_\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Methods inherited from sklearn.base.ClassifierMixin:\n",
|
||
|
|
" | \n",
|
||
|
|
" | score(self, X, y, sample_weight=None)\n",
|
||
|
|
" | Return the mean accuracy on the given test data and labels.\n",
|
||
|
|
" | \n",
|
||
|
|
" | In multi-label classification, this is the subset accuracy\n",
|
||
|
|
" | which is a harsh metric since you require for each sample that\n",
|
||
|
|
" | each label set be correctly predicted.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : array-like of shape (n_samples, n_features)\n",
|
||
|
|
" | Test samples.\n",
|
||
|
|
" | \n",
|
||
|
|
" | y : array-like of shape (n_samples,) or (n_samples, n_outputs)\n",
|
||
|
|
" | True labels for X.\n",
|
||
|
|
" | \n",
|
||
|
|
" | sample_weight : array-like of shape (n_samples,), default=None\n",
|
||
|
|
" | Sample weights.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | score : float\n",
|
||
|
|
" | Mean accuracy of self.predict(X) wrt. y.\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Data descriptors inherited from sklearn.base.ClassifierMixin:\n",
|
||
|
|
" | \n",
|
||
|
|
" | __dict__\n",
|
||
|
|
" | dictionary for instance variables (if defined)\n",
|
||
|
|
" | \n",
|
||
|
|
" | __weakref__\n",
|
||
|
|
" | list of weak references to the object (if defined)\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Methods inherited from sklearn.svm._base.BaseLibSVM:\n",
|
||
|
|
" | \n",
|
||
|
|
" | fit(self, X, y, sample_weight=None)\n",
|
||
|
|
" | Fit the SVM model according to the given training data.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | X : {array-like, sparse matrix} of shape (n_samples, n_features) or (n_samples, n_samples)\n",
|
||
|
|
" | Training vectors, where n_samples is the number of samples\n",
|
||
|
|
" | and n_features is the number of features.\n",
|
||
|
|
" | For kernel=\"precomputed\", the expected shape of X is\n",
|
||
|
|
" | (n_samples, n_samples).\n",
|
||
|
|
" | \n",
|
||
|
|
" | y : array-like of shape (n_samples,)\n",
|
||
|
|
" | Target values (class labels in classification, real numbers in\n",
|
||
|
|
" | regression)\n",
|
||
|
|
" | \n",
|
||
|
|
" | sample_weight : array-like of shape (n_samples,), default=None\n",
|
||
|
|
" | Per-sample weights. Rescale C per sample. Higher weights\n",
|
||
|
|
" | force the classifier to put more emphasis on these points.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | self : object\n",
|
||
|
|
" | \n",
|
||
|
|
" | Notes\n",
|
||
|
|
" | -----\n",
|
||
|
|
" | If X and y are not C-ordered and contiguous arrays of np.float64 and\n",
|
||
|
|
" | X is not a scipy.sparse.csr_matrix, X and/or y may be copied.\n",
|
||
|
|
" | \n",
|
||
|
|
" | If X is a dense array, then the other methods will not support sparse\n",
|
||
|
|
" | matrices as input.\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Data descriptors inherited from sklearn.svm._base.BaseLibSVM:\n",
|
||
|
|
" | \n",
|
||
|
|
" | coef_\n",
|
||
|
|
" | \n",
|
||
|
|
" | n_support_\n",
|
||
|
|
" | \n",
|
||
|
|
" | ----------------------------------------------------------------------\n",
|
||
|
|
" | Methods inherited from sklearn.base.BaseEstimator:\n",
|
||
|
|
" | \n",
|
||
|
|
" | __getstate__(self)\n",
|
||
|
|
" | \n",
|
||
|
|
" | __repr__(self, N_CHAR_MAX=700)\n",
|
||
|
|
" | Return repr(self).\n",
|
||
|
|
" | \n",
|
||
|
|
" | __setstate__(self, state)\n",
|
||
|
|
" | \n",
|
||
|
|
" | get_params(self, deep=True)\n",
|
||
|
|
" | Get parameters for this estimator.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | deep : bool, default=True\n",
|
||
|
|
" | If True, will return the parameters for this estimator and\n",
|
||
|
|
" | contained subobjects that are estimators.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | params : mapping of string to any\n",
|
||
|
|
" | Parameter names mapped to their values.\n",
|
||
|
|
" | \n",
|
||
|
|
" | set_params(self, **params)\n",
|
||
|
|
" | Set the parameters of this estimator.\n",
|
||
|
|
" | \n",
|
||
|
|
" | The method works on simple estimators as well as on nested objects\n",
|
||
|
|
" | (such as pipelines). The latter have parameters of the form\n",
|
||
|
|
" | ``<component>__<parameter>`` so that it's possible to update each\n",
|
||
|
|
" | component of a nested object.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Parameters\n",
|
||
|
|
" | ----------\n",
|
||
|
|
" | **params : dict\n",
|
||
|
|
" | Estimator parameters.\n",
|
||
|
|
" | \n",
|
||
|
|
" | Returns\n",
|
||
|
|
" | -------\n",
|
||
|
|
" | self : object\n",
|
||
|
|
" | Estimator instance.\n",
|
||
|
|
"\n"
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"help(SVC)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"**NOTE: For this example, we will explore the algorithm, so we'll skip any scaling or even a train\\test split for now**"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 9,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"y = df['Virus Present']\n",
|
||
|
|
"X = df.drop('Virus Present',axis=1) "
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 10,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"SVC(C=1000, kernel='linear')"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 10,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='linear', C=1000)\n",
|
||
|
|
"model.fit(X, y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 11,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"# This is imported from the supplemental .py file\n",
|
||
|
|
"# https://scikit-learn.org/stable/auto_examples/svm/plot_separating_hyperplane.html\n",
|
||
|
|
"from svm_margin_plot import plot_svm_boundary"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 12,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## Hyper Parameters\n",
|
||
|
|
"\n",
|
||
|
|
"### C\n",
|
||
|
|
"\n",
|
||
|
|
"Regularization parameter. The strength of the regularization is **inversely** proportional to C. Must be strictly positive. The penalty is a squared l2 penalty.\n",
|
||
|
|
"\n",
|
||
|
|
"*Note: If you are following along with the equations, specifically the value of C as described in ISLR, C in scikit-learn is **inversely** proportional to this value.*"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 13,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"SVC(C=0.05, kernel='linear')"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 13,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='linear', C=0.05)\n",
|
||
|
|
"model.fit(X, y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 14,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"### Kernel\n",
|
||
|
|
"\n",
|
||
|
|
"[Choosing a Kernel](https://stats.stackexchange.com/questions/18030/how-to-select-kernel-for-svm?rq=1)\n",
|
||
|
|
"\n",
|
||
|
|
"#### rbf - [Radial Basis Function](https://en.wikipedia.org/wiki/Radial_basis_function_kernel)\n",
|
||
|
|
"\n",
|
||
|
|
"\n",
|
||
|
|
"When training an SVM with the Radial Basis Function (RBF) kernel, two parameters must be considered: C and gamma. The parameter C, common to all SVM kernels, trades off misclassification of training examples against simplicity of the decision surface. A low C makes the decision surface smooth, while a high C aims at classifying all training examples correctly. gamma defines how much influence a single training example has. The larger gamma is, the closer other examples must be to be affected."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 15,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='rbf', C=1)\n",
|
||
|
|
"model.fit(X, y)\n",
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 16,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='sigmoid')\n",
|
||
|
|
"model.fit(X, y)\n",
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"#### Degree (poly kernels only)\n",
|
||
|
|
"\n",
|
||
|
|
"Degree of the polynomial kernel function ('poly').\n",
|
||
|
|
"Ignored by all other kernels."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 17,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='poly', C=1,degree=1)\n",
|
||
|
|
"model.fit(X, y)\n",
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 18,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='poly', C=1,degree=2)\n",
|
||
|
|
"model.fit(X, y)\n",
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"### gamma\n",
|
||
|
|
"\n",
|
||
|
|
"gamma : {'scale', 'auto'} or float, default='scale'\n",
|
||
|
|
" Kernel coefficient for 'rbf', 'poly' and 'sigmoid'.\n",
|
||
|
|
"\n",
|
||
|
|
" - if ``gamma='scale'`` (default) is passed then it uses\n",
|
||
|
|
" 1 / (n_features * X.var()) as value of gamma,\n",
|
||
|
|
" - if 'auto', uses 1 / n_features."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 19,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"image/png": "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
|
||
|
|
"text/plain": [
|
||
|
|
"<Figure size 432x288 with 1 Axes>"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"metadata": {
|
||
|
|
"needs_background": "light"
|
||
|
|
},
|
||
|
|
"output_type": "display_data"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"model = SVC(kernel='rbf', C=1,gamma=0.01)\n",
|
||
|
|
"model.fit(X, y)\n",
|
||
|
|
"plot_svm_boundary(model,X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"## Grid Search\n",
|
||
|
|
"\n",
|
||
|
|
"Keep in mind, for this simple example, we saw the classes were easily separated, which means each variation of model could easily get 100% accuracy, meaning a grid search is \"useless\"."
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 20,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"from sklearn.model_selection import GridSearchCV"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 21,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [],
|
||
|
|
"source": [
|
||
|
|
"svm = SVC()\n",
|
||
|
|
"param_grid = {'C':[0.01,0.1,1],'kernel':['linear','rbf']}\n",
|
||
|
|
"grid = GridSearchCV(svm,param_grid)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 22,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"GridSearchCV(estimator=SVC(),\n",
|
||
|
|
" param_grid={'C': [0.01, 0.1, 1, 10], 'kernel': ['linear', 'rbf']})"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 22,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"# Note again we didn't split Train|Test\n",
|
||
|
|
"grid.fit(X,y)"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 23,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"1.0"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 23,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"# 100% accuracy (as expected)\n",
|
||
|
|
"grid.best_score_"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "code",
|
||
|
|
"execution_count": 24,
|
||
|
|
"metadata": {},
|
||
|
|
"outputs": [
|
||
|
|
{
|
||
|
|
"data": {
|
||
|
|
"text/plain": [
|
||
|
|
"{'C': 0.01, 'kernel': 'linear'}"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
"execution_count": 24,
|
||
|
|
"metadata": {},
|
||
|
|
"output_type": "execute_result"
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"source": [
|
||
|
|
"grid.best_params_"
|
||
|
|
]
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"cell_type": "markdown",
|
||
|
|
"metadata": {},
|
||
|
|
"source": [
|
||
|
|
"This is more to review the grid search process, recall in a real situation such as your exercise, you will perform a train|test split and get final evaluation metrics."
|
||
|
|
]
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"metadata": {
|
||
|
|
"anaconda-cloud": {},
|
||
|
|
"kernelspec": {
|
||
|
|
"display_name": "Python 3",
|
||
|
|
"language": "python",
|
||
|
|
"name": "python3"
|
||
|
|
},
|
||
|
|
"language_info": {
|
||
|
|
"codemirror_mode": {
|
||
|
|
"name": "ipython",
|
||
|
|
"version": 3
|
||
|
|
},
|
||
|
|
"file_extension": ".py",
|
||
|
|
"mimetype": "text/x-python",
|
||
|
|
"name": "python",
|
||
|
|
"nbconvert_exporter": "python",
|
||
|
|
"pygments_lexer": "ipython3",
|
||
|
|
"version": "3.8.5"
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"nbformat": 4,
|
||
|
|
"nbformat_minor": 1
|
||
|
|
}
|