cas-pml/SL/aufgaben/WS1/notebooks/2.4 Klassifikation - Neuronale Netze.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Feature Engineering\n",
    "# Klassifikation\n",
    "## Instanzbasierte Modelle\n",
    "## Regelbasierte Modelle\n",
    "## Mathematische Modelle\n",
    "## Neuronale Netze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('./')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-17T12:01:39.858981Z",
     "start_time": "2020-03-17T12:01:37.904657Z"
    }
   },
   "outputs": [],
   "source": [
    "## preparation\n",
    "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)\n",
    "\n",
    "from bfh_cas_pml import prep_data, prep_demo_data\n",
    "X_train, X_test, y_train, y_test = prep_data('bank_data_prep.csv', 'y', seed = 1234)\n",
    "X_demo, y_demo = prep_demo_data('demo_data_class.csv', 'y')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### MLPClassifier\n",
    "#### Theorie"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "kein Code zu diesem Kapitel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Praxis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-17T12:01:51.083863Z",
     "start_time": "2020-03-17T12:01:48.211064Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6729540614542135\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neural_network import MLPClassifier\n",
    "model = MLPClassifier(random_state = 1234)\n",
    "model.fit(X_train, y_train) \n",
    "print(model.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-17T12:01:51.116594Z",
     "start_time": "2020-03-17T12:01:51.088588Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'activation': 'relu', 'alpha': 0.0001, 'batch_size': 'auto', 'beta_1': 0.9, 'beta_2': 0.999, 'early_stopping': False, 'epsilon': 1e-08, 'hidden_layer_sizes': (100,), 'learning_rate': 'constant', 'learning_rate_init': 0.001, 'max_fun': 15000, 'max_iter': 200, 'momentum': 0.9, 'n_iter_no_change': 10, 'nesterovs_momentum': True, 'power_t': 0.5, 'random_state': 1234, 'shuffle': True, 'solver': 'adam', 'tol': 0.0001, 'validation_fraction': 0.1, 'verbose': False, 'warm_start': False}\n"
     ]
    }
   ],
   "source": [
    "print(model.get_params())"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "## show weights\n",
    "print(model.intercepts_)\n",
    "print(model.coefs_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-03-17T12:06:20.818868Z",
     "start_time": "2020-03-17T12:06:20.800859Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 0.6729540614542135\n",
      "1 0.5308792211743231\n",
      "2 0.705506540918771\n",
      "3 0.7222391238211134\n",
      "4 0.7751749315485245\n",
      "5 0.7121995740797079\n",
      "6 0.6759963492546395\n",
      "7 0.48098570124733797\n",
      "8 0.6096744752053544\n",
      "9 0.6638271980529358\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## investigate stability of MLPClassifier using different random seeds\n",
    "params = range(0, 10)\n",
    "scores = []\n",
    "\n",
    "for param in params:\n",
    "    model = MLPClassifier(random_state = 1234 + param)\n",
    "    model.fit(X_train, y_train) \n",
    "    score = model.score(X_test, y_test)\n",
    "    scores.append(score)\n",
    "    print(param, score)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(2,4))\n",
    "sns.boxplot(y=scores)\n",
    "ax.set(ylabel='accuracy');"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "teaching",
   "language": "python",
   "name": "teaching"
  },
  "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": "2.4",
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "2.4 Klassifikation - Neuronale Netze",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "165px"
   },
   "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": "316.717px",
    "left": "782px",
    "right": "20px",
    "top": "119px",
    "width": "350px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}