{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"tags": []
},
"source": [
"# Feature Engineering\n",
"# Klassifikation\n",
"## Instanzbasierte Modelle"
]
},
{
"cell_type": "markdown",
"metadata": {
"tags": []
},
"source": [
"**Vorbereitung der Umgebung**"
]
},
{
"cell_type": "raw",
"metadata": {
"tags": []
},
"source": [
"## for scikit-learn 1.4.2, to silence warnings regarding physical cores\n",
"import os\n",
"os.environ['LOKY_MAX_CPU_COUNT'] = '4' ## depending on the hardware used"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns; sns.set()\n",
"%matplotlib inline\n",
"\n",
"datapath = '../3_data'\n",
"from os import chdir; chdir(datapath)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Vorbereitung der Daten**"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"## load data\n",
"data = pd.read_csv('bank_data_prep.csv')\n",
"data.shape ## check\n",
"\n",
"## features - target - split\n",
"## organize features and target as independent objects\n",
"X = data.drop('y', axis=1)\n",
"y = data['y']\n",
"\n",
"## test - train - split\n",
"from sklearn.model_selection import train_test_split\n",
"X_train, X_test, y_train, y_test = train_test_split(\n",
" X,\n",
" y,\n",
" train_size=2 / 3,\n",
" random_state=1234)\n",
"\n",
"## demo dataset\n",
"demo_data = pd.read_csv('demo_data_class.csv')\n",
"X_demo = demo_data.drop('y', axis=1)\n",
"y_demo = demo_data['y']"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### KNeighborsClassifier\n",
"#### Theorie\n",
"#### Praxis"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"ExecuteTime": {
"end_time": "2020-03-17T11:35:53.227097Z",
"start_time": "2020-03-17T11:35:53.124006Z"
}
},
"outputs": [],
"source": [
"## import trainer class\n",
"from sklearn.neighbors import KNeighborsClassifier"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"ExecuteTime": {
"end_time": "2020-03-17T11:35:53.242354Z",
"start_time": "2020-03-17T11:35:53.231829Z"
}
},
"outputs": [],
"source": [
"## instantiate (and parameterize) the model\n",
"model = KNeighborsClassifier()"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"ExecuteTime": {
"end_time": "2020-03-17T11:35:53.389975Z",
"start_time": "2020-03-17T11:35:53.247879Z"
},
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"KNeighborsClassifier() In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"%%time\n",
"\n",
"## another outlook: parameter tuning\n",
"from sklearn.neighbors import KNeighborsClassifier\n",
"model = KNeighborsClassifier()\n",
"\n",
"params = range(1, 21) ## k values as range from 1 to 20 by 1\n",
"scores = [] ## empty list for collecting score results by iteration\n",
"\n",
"## iterate over params\n",
"for param in params:\n",
" model.set_params(n_neighbors=param)\n",
" model.fit(X_train, y_train)\n",
" scores.append(model.score(X_test, y_test))\n",
" print(param, model.score(X_test, y_test)) ## for trace progress only\n",
" \n",
"## visualization\n",
"fig = sns.lineplot(x=params, y=scores)\n",
"plt.scatter(x=params[scores.index(max(scores))], y=max(scores), color=\"black\")\n",
"plt.xlabel('k')\n",
"plt.ylabel('accuracy');"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Standardisieren von gesplitteten Daten"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"ExecuteTime": {
"end_time": "2020-03-17T11:36:12.816582Z",
"start_time": "2020-03-17T11:36:12.761181Z"
}
},
"outputs": [],
"source": [
"## get standardization parameters on the base of training data\n",
"from sklearn.preprocessing import StandardScaler\n",
"scaler = StandardScaler().set_output(transform='pandas')\n",
"scaler.fit(X_train)\n",
"\n",
"## apply standardization to both data sets\n",
"X_train_sc = scaler.transform(X_train)\n",
"X_test_sc = scaler.transform(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"ExecuteTime": {
"end_time": "2020-03-17T11:36:12.816582Z",
"start_time": "2020-03-17T11:36:12.761181Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[40.2715655 4.4179218 0.55256352]\n",
"age 40.271565\n",
"education 4.417922\n",
"housing 0.552564\n",
"dtype: float64\n",
"[11.83482251 1.65671617 0.4972294 ]\n",
"age 11.835723\n",
"education 1.656842\n",
"housing 0.497267\n",
"dtype: float64\n",
"age 1.297203e-17\n",
"education -1.086407e-16\n",
"housing 1.135052e-16\n",
"dtype: float64\n",
"age 1.000076\n",
"education 1.000076\n",
"housing 1.000076\n",
"dtype: float64\n",
"age 0.027978\n",
"education 0.023008\n",
"housing 0.001667\n",
"dtype: float64\n",
"age 0.995850\n",
"education 0.986509\n",
"housing 0.999975\n",
"dtype: float64\n"
]
}
],
"source": [
"## check\n",
"## compare mean and std of data and scaler attribute\n",
"print(scaler.mean_[0:3])\n",
"print(X_train.iloc[:,[0,1,2]].mean())\n",
"print(scaler.scale_[0:3])\n",
"print(X_train.iloc[:,[0,1,2]].std())\n",
"\n",
"## compare mean and std before and after scalint\n",
"print(X_train_sc.iloc[:,[0,1,2]].mean())\n",
"print(X_train_sc.iloc[:,[0,1,2]].std())\n",
"print(X_test_sc.iloc[:,[0,1,2]].mean())\n",
"print(X_test_sc.iloc[:,[0,1,2]].std())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Weitere Instanzbasierte Methoden"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"(kein Code)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Data Preparation für weitere Klassifikatoren (und Regressoren)"
]
},
{
"cell_type": "raw",
"metadata": {},
"source": [
"## read and prep real data\n",
"from bfh_cas_pml import prep_data\n",
"X_train, X_test, y_train, y_test = prep_data(\n",
" 'bank_data_prep.csv', 'y', seed = 1234)"
]
},
{
"cell_type": "raw",
"metadata": {},
"source": [
"## read and prep demo data\n",
"from bfh_cas_pml prep_demo_data\n",
"X_demo, y_demo = prep_demo_data('demo_data_class.csv', 'y')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"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": "3.13.0"
},
"toc": {
"base_numbering": "2.1",
"nav_menu": {},
"number_sections": true,
"sideBar": true,
"skip_h1_title": false,
"title_cell": "Inhalt",
"title_sidebar": "Inhalt",
"toc_cell": true,
"toc_position": {
"height": "calc(100% - 180px)",
"left": "10px",
"top": "150px",
"width": "173.533px"
},
"toc_section_display": true,
"toc_window_display": true
},
"toc-autonumbering": true,
"varInspector": {
"cols": {
"lenName": 16,
"lenType": 16,
"lenVar": 40
},
"kernels_config": {
"python": {
"delete_cmd_postfix": "",
"delete_cmd_prefix": "del ",
"library": "var_list.py",
"varRefreshCmd": "print(var_dic_list())"
},
"r": {
"delete_cmd_postfix": ") ",
"delete_cmd_prefix": "rm(",
"library": "var_list.r",
"varRefreshCmd": "cat(var_dic_list()) "
}
},
"position": {
"height": "144.183px",
"left": "910px",
"right": "20px",
"top": "115px",
"width": "350px"
},
"types_to_exclude": [
"module",
"function",
"builtin_function_or_method",
"instance",
"_Feature"
],
"window_display": false
}
},
"nbformat": 4,
"nbformat_minor": 4
}