{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Store and Reload the Trained Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This example shows how to store a trained hyperbox-based model and reload it to make prediction. This example will use a random hyperboxes model for illustration."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_classification\n",
    "from hbbrain.numerical_data.incremental_learner.iol_gfmm import ImprovedOnlineGFMM\n",
    "from hbbrain.numerical_data.ensemble_learner.random_hyperboxes import RandomHyperboxesClassifier\n",
    "from hbbrain.utils.model_storage import store_model, load_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = make_classification(n_samples=100, n_features=4, n_informative=2, n_redundant=0, random_state=0, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Normalise data into the range of [0, 1]\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "scaler = MinMaxScaler()\n",
    "scaler.fit(X)\n",
    "X = scaler.transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training a random hyperboxes model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = RandomHyperboxesClassifier(base_estimator=ImprovedOnlineGFMM(0.1), n_estimators=10, random_state=0).fit(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted class for the input patter [1, 0.6, 0.5, 0.2] is 1\n"
     ]
    }
   ],
   "source": [
    "y_pred = clf.predict([[1, 0.6, 0.5, 0.2]])\n",
    "print(\"Predicted class for the input patter [1, 0.6, 0.5, 0.2] is %d\"%y_pred[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Store the trained model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "store_model(clf, \"store_example_model.dummy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reload the trained model and make prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_load = load_model(\"store_example_model.dummy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted class for the input patter [1, 0.6, 0.5, 0.2] is 1\n"
     ]
    }
   ],
   "source": [
    "y_pred = clf_load.predict([[1, 0.6, 0.5, 0.2]])\n",
    "print(\"Predicted class for the input patter [1, 0.6, 0.5, 0.2] is %d\"%y_pred[0])"
   ]
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}