Module-5/week-1/day-2/knn/knn_classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# K Nearest Neighbor\n",
    "## A movie-ing target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Friend inventory\n",
    "### Use handout\n",
    "\n",
    "![bullseye3](img/bullseye2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Friend Inventory\n",
    "### What would each person choose?\n",
    "\n",
    "![movies](img/movies.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Friend Inventory\n",
    "### Decision majority by ring\n",
    "\n",
    "If you just polled the inner ring of people, what movie would you end up seeing?\n",
    "- How about if you polled the first *and* second ring?\n",
    "- The first three rings?\n",
    "- All of it?\n",
    "\n",
    "Share with your neighbor what movie you'd end up seeing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Friend Inventory\n",
    "### What did you end up seeing?\n",
    "\n",
    "Who's movie choices changed based on how many people you polled?\n",
    "![movies](img/movies.png)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Friend Inventory\n",
    "### What's the \"algorithm\" we used for this process, in normal words?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "### How does this relate to K nearest neighbor(knn)?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "![annotate](img/bullseye-annotate.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Starting Off\n",
    "\n",
    "We are going to revisit the pima native Americans diabetes dataset. Can we use knn to classify people correctly and then predict if someone will have diabetes?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Implementing a K-Nearest Neighbors Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Objective: \n",
    "- To be able to implement a KNN classification model, and utilize a visualization to identify the best value of K."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Agenda\n",
    "\n",
    "- What is the **K-nearest neighbors** classification model?\n",
    "- How do you use the sklearn grid search function?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## K-nearest neighbors (KNN) classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "**Lazy learning** or **Instance-based (IB)** learning methods simply store the training examples and postpone the generalization (building a model) until a new instance must be classified or prediction made."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "**Non-parametric models** assume that the data distribution cannot be defined in\n",
    "terms of such a finite set of parameters."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How does the KNN algorithm work?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "### What should the grey point be?\n",
    "\n",
    "<img src='img/scenario.png' width=500/>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## KNN has the following basic steps:\n",
    "\n",
    "<img src='img/knn-process.png' width=500/>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Voting\n",
    "\n",
    "How to break ties:\n",
    "\n",
    "1. When doing a binary classification, often use a odd K to avoid ties.\n",
    "2. Multiple approaches for Multiclass problems:\n",
    "    - Reduce the K by 1 to see who wins.\n",
    "    - Weight the votes based on the distance of the neighbors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Example training data\n",
    "\n",
    "This example uses a multi-class problem and each color represents a different class. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### KNN classification map (K=1)\n",
    "\n",
    "![1NN classification map](img/04_1nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### KNN classification map (K=5)\n",
    "\n",
    "![5NN classification map](img/04_5nn_map.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A bit more math\n",
    "\n",
    "### Euclidean Distance\n",
    "\n",
    "**Euclidean distance** refers to the distance between two points. These points can be in different dimensional space and are represented by different forms of coordinates. In one-dimensional space, the points are just on a straight number line.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A bit more math\n",
    "### Measuring distance in a 2-d Space\n",
    "\n",
    "In two-dimensional space, the coordinates are given as points on the x- and y-axes\n",
    "\n",
    "![alt text](img/euclidean-distance.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A bit more math\n",
    "### Measuring distance in a 3-d Space\n",
    "\n",
    "In three-dimensional space, x-, y- and z-axes are used. \n",
    "\n",
    "$$\\sqrt{(x_1-x_2)^2 + (y_1-y_2)^2 +  (z_1-z_2)^2}$$\n",
    "![alt text](img/vectorgraph.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## A bit more math\n",
    "### Euclidean Distance Equation\n",
    "![alt text](img/euclidean-equation.png)\n",
    "\n",
    "The source of this formula is in the Pythagorean theorem. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Implementing the KNN Classifier with SKlearn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Reviewing the Pima dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Loading the data\n",
    "\n",
    "\n",
    "[following this analysis for part of this section](https://www.kaggle.com/shrutimechlearn/step-by-step-diabetes-classification-knn-detailed)\n",
    "\n",
    "The Question creeping out of this summary\n",
    "- Can minimum value of below listed columns be zero (0)?\n",
    "- On these columns, a value of zero does not make sense and thus indicates missing value.\n",
    "\n",
    "Following columns or variables have an invalid zero value:\n",
    "\n",
    "- Glucose\n",
    "- BloodPressure\n",
    "- SkinThickness\n",
    "- Insulin\n",
    "- BMI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "diabetes = pd.read_csv('diabetes.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# print the shapes of X and y\n",
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# Use x and y variables to split the training data into train and test set\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Importance of Scaling\n",
    "\n",
    "Compare how the different the data looks when it is scaled versus non-scaled\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Should we use a Standard Scaler or Min-Max Scaler?\n",
    "\n",
    "https://sebastianraschka.com/Articles/2014_about_feature_scaling.html\n",
    "http://datareality.blogspot.com/2016/11/scaling-normalizing-standardizing-which.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/swilson5/anaconda3/lib/python3.6/site-packages/sklearn/preprocessing/data.py:625: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "/Users/swilson5/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:7: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  import sys\n",
      "/Users/swilson5/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:8: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "scaler = StandardScaler()  \n",
    "scaler.fit(X_train)\n",
    "\n",
    "X_train = scaler.transform(X_train)  \n",
    "X_test = scaler.transform(X_test)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## scikit-learn 4-step modeling pattern"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "![steps](img/sklearnsteps.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Step 1:** Import the class you plan to use"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Step 2:** \"Instantiate\" the \"estimator\"\n",
    "\n",
    "- \"Estimator\" is scikit-learn's term for model\n",
    "- \"Instantiate\" means \"make an instance of\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(n_neighbors=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Class specifications**\n",
    "- Name of the object does not matter\n",
    "- Can specify tuning parameters (aka \"hyperparameters\") during this step\n",
    "- All parameters not specified are set to their defaults"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_jobs=None, n_neighbors=1, p=2,\n",
      "           weights='uniform')\n"
     ]
    }
   ],
   "source": [
    "print(knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Step 3:** Fit the model with data (aka \"model training\")\n",
    "\n",
    "- Model is learning the relationship between X and y\n",
    "- Occurs in-place"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "           metric_params=None, n_jobs=None, n_neighbors=1, p=2,\n",
       "           weights='uniform')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "**Step 4:** Predict the response for a new observation\n",
    "\n",
    "- New observations are called \"out-of-sample\" data\n",
    "- Uses the information it learned during the model training process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "# make class predictions for the testing set\n",
    "y_pred_class = knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# calculate accuracy\n",
    "from sklearn import metrics\n",
    "print('Accuracy:' + str(metrics.accuracy_score(y_test, y_pred_class)))\n",
    "print('F1: ' + str(metrics.f1_score(y_test, y_pred_class)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import itertools\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "plt.rcParams[\"figure.figsize\"] = [10,5]\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion Matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test,y_pred_class)\n",
    "classes = ['Perished', 'Survived']\n",
    "plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Using a different value for K"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# instantiate the model (using the value K=5)\n",
    "knn = KNeighborsClassifier(n_neighbors=5)\n",
    "\n",
    "# fit the model with data\n",
    "knn.fit(X_train, y_train)\n",
    "\n",
    "# make class predictions for the testing set\n",
    "y_pred_class = knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "print('Accuracy:' + str(metrics.accuracy_score(y_test, y_pred_class)))\n",
    "print('F1: ' + str(metrics.f1_score(y_test, y_pred_class)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report  \n",
    "\n",
    "print(classification_report(y_test, y_pred_class))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm = confusion_matrix(y_test,y_pred_class)\n",
    "classes = ['Perished', 'Survived']\n",
    "plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Search for an optimal value of K for KNN\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "k_range = list(range(1, 4))\n",
    "k_scores = []\n",
    "for k in k_range:\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    knn.fit(X_train, y_train)\n",
    "    y_predict = knn.predict(X_test)\n",
    "    score = metrics.f1_score(y_test, y_predict, average='weighted')\n",
    "    k_scores.append( score)\n",
    "print(k_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Visual comparison of different $K$s\n",
    "\n",
    "This is not an ROC curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt  \n",
    "\n",
    "plt.figure(figsize=(12, 6))  \n",
    "plt.plot(k_range, k_scores, color='red', linestyle='dashed', marker='o',  \n",
    "         markerfacecolor='blue', markersize=10)\n",
    "plt.title('F1 score by K Value')  \n",
    "plt.xlabel('K Value')  \n",
    "plt.ylabel('F1 Score') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### What value of K performs best on our Test data?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Here we use F score, what other metrics could we use?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How do you think K size relates to our concepts of bias and variance?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![alt text](img/K-NN_Neighborhood_Size_print.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Resources\n",
    "\n",
    "- [Nearest Neighbors](http://scikit-learn.org/stable/modules/neighbors.html) (user guide), [KNeighborsClassifier](http://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html) (class documentation)\n",
    "\n",
    "- [Videos from An Introduction to Statistical Learning](http://www.dataschool.io/15-hours-of-expert-machine-learning-videos/)\n",
    "    - Classification Problems and K-Nearest Neighbors (Chapter 2)\n",
    "    - Introduction to Classification (Chapter 4)\n",
    "    - Logistic Regression and Maximum Likelihood (Chapter 4)"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}