Morten Hjorth-Jensen
126 lines
22 KiB
Plaintext
126 lines
22 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"(506, 104)\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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\n",
|
|
"text/plain": [
|
|
"<Figure size 432x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"training set score: 0.771865\n",
|
|
"test set score: 0.725968\n",
|
|
"training set score: 0.771686\n",
|
|
"test set score: 0.722415\n",
|
|
"training set score: 0.771863\n",
|
|
"test set score: 0.725626\n",
|
|
"--------------------\n",
|
|
"training set score: 0.771865\n",
|
|
"test set score: 0.725916\n",
|
|
"number of features used: 2\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"#!pip install mglearn\n",
|
|
"import mglearn\n",
|
|
"import sklearn\n",
|
|
"import pandas as pd\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"import numpy as np\n",
|
|
"import IPython\n",
|
|
"\n",
|
|
"from sklearn.datasets import load_boston\n",
|
|
"boston = load_boston()\n",
|
|
"X, y = mglearn.datasets.load_extended_boston()\n",
|
|
"print(X.shape)\n",
|
|
"mglearn.plots.plot_knn_classification(n_neighbors=3)\n",
|
|
"plt.show()\n",
|
|
"\n",
|
|
"from sklearn.model_selection import train_test_split\n",
|
|
"X, y=mglearn.datasets.make_forge()\n",
|
|
"\n",
|
|
"X_train, X_test, y_train, y_test=train_test_split(X, y, random_state=0)\n",
|
|
"\n",
|
|
"from sklearn.neighbors import KNeighborsClassifier\n",
|
|
"clf=KNeighborsClassifier(n_neighbors=3)\n",
|
|
"clf.fit(X_train, y_train)\n",
|
|
"KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski')\n",
|
|
"clf.predict(X_test)\n",
|
|
"clf.score(X_test, y_test)\n",
|
|
"\n",
|
|
"from sklearn.linear_model import LinearRegression\n",
|
|
"lr=LinearRegression().fit(X_train, y_train)\n",
|
|
"\n",
|
|
"print(\"training set score: %f\" % lr.score(X_train, y_train))\n",
|
|
"print(\"test set score: %f\" % lr.score(X_test, y_test))\n",
|
|
"\n",
|
|
"from sklearn.linear_model import Ridge\n",
|
|
"ridge = Ridge().fit(X_train, y_train)\n",
|
|
"print(\"training set score: %f\" % ridge.score(X_train, y_train))\n",
|
|
"print(\"test set score: %f\" % ridge.score(X_test, y_test))\n",
|
|
"\n",
|
|
"ridge01 = Ridge(alpha=0.1).fit(X_train, y_train)\n",
|
|
"print(\"training set score: %f\" % ridge01.score(X_train, y_train))\n",
|
|
"print(\"test set score: %f\" % ridge01.score(X_test, y_test))\n",
|
|
"\n",
|
|
"print (\"--------------------\")\n",
|
|
"\n",
|
|
"from sklearn.linear_model import Lasso\n",
|
|
"lasso00001 = Lasso(alpha=0.0001).fit(X_train, y_train)\n",
|
|
"print(\"training set score: %f\" % lasso00001.score(X_train, y_train))\n",
|
|
"print(\"test set score: %f\" % lasso00001.score(X_test, y_test))\n",
|
|
"print(\"number of features used: %d\" % np.sum(lasso00001.coef_ != 0))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"collapsed": true
|
|
},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"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": "3.7.0"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|