{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "cell_style": "center",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "![CDS 411 logo](../../img/cds-411-logo.png)\n",
    "\n",
    "# Class 15: Data-driven modeling II\n",
    "\n",
    "---\n",
    "\n",
    "![CC BY-SA 4.0 license](../../img/cc-by-sa.png)\n",
    "\n",
    "This notebook is licensed under a [Creative Commons Attribution-ShareAlike 4.0 International License](http://creativecommons.org/licenses/by-sa/4.0/)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "cell_style": "center",
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Load packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "cell_style": "center",
    "slideshow": {
     "cell_style": "split",
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import cross_val_score, RepeatedKFold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Select a random seed for notebook reproducibility."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(862604190)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The *Filip* dataset\n",
    "\n",
    "Source: <https://www.itl.nist.gov/div898/strd/lls/data/Filip.shtml>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Last class we took a look at the *Filip* dataset. The model that generated this dataset was a polynomial summation, $y=\\beta_{0}+\\beta_{1}x+\\beta_{2}x^{2}+\\beta_{3}x^{3}+\\dots{}$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "filip_csv_path = Path(\"../../data/nist/filip.csv\")\n",
    "filip_df = pd.read_csv(filip_csv_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data looks as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(dpi=90)\n",
    "sns.scatterplot(x=\"x\", y=\"y\", data=filip_df, ax=ax);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Last time, we used the `statsmodels` package to fit the formula $y=\\beta_{0}+\\beta_{1}x+\\beta_{2}x^2+\\beta_{3}x^3$ to the data.\n",
    "\n",
    "I glossed over the details underpinning how the `statsmodels` package worked, because, while it is a useful package to know how to use, it is not the main one I want to use. Instead, we need to learn how to perform the same type of fit using the powerful `scikit-learn` package."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is `scikit-learn`?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[scikit-learn](http://scikit-learn.org) is the premiere machine learning package for Python, similar to what [caret](https://topepo.github.io/caret/) is for the R language."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![scikit-learn front page](../../img/scikit-learn-frontpage.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you use Python and plan to do something with machine learning, you **will** be using this package. It's worth learning how to navigate using it, even for simple tools such as linear regression."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So what kinds of problems should you use it for? The `scikit-learn` documentation provides [a nice overview of this](http://scikit-learn.org/stable/tutorial/basic/tutorial.html#machine-learning-the-problem-setting):\n",
    "\n",
    "> In general, a learning problem considers a set of n [samples](https://en.wikipedia.org/wiki/Sample_(statistics)) of data and then tries to predict properties of unknown data. If each sample is more than a single number and, for instance, a multi-dimensional entry (aka [multivariate](https://en.wikipedia.org/wiki/Multivariate_random_variable) data), it is said to have several attributes or **features**.\n",
    ">\n",
    "> Learning problems fall into a few categories:\n",
    ">\n",
    "> *   [supervised learning](https://en.wikipedia.org/wiki/Supervised_learning), in which the data comes with additional attributes that we want to predict \\[...\\] This problem can be either:\n",
    ">\n",
    ">     *    [classification](https://en.wikipedia.org/wiki/Classification_in_machine_learning): samples belong to two or more classes and we want to learn from already labeled data how to predict the class of unlabeled data.\n",
    ">          \\[...\\]\n",
    "> \n",
    ">     *    [regression](https://en.wikipedia.org/wiki/Regression_analysis): if the desired output consists of one or more continuous variables, then the task is called regression.\n",
    ">          \\[...\\]\n",
    "> \n",
    "> *   [unsupervised learning](https://en.wikipedia.org/wiki/Unsupervised_learning), in which the training data consists of a set of input vectors x without any corresponding target values."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is in the package?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A lot! Many of the steps to fitting any given model follow a similar workflow, so it's been possible to bring many models and techniques to bear under the single umbrella of `scikit-learn`. Just a quick sampling of the [User Guide](http://scikit-learn.org/stable/user_guide.html) table of contents gives you an idea of what's available:\n",
    "\n",
    "1.  **Supervised learning**\n",
    "    *   Models classes include: generalized linear models, kernel ridge regression, support vector machines, naive bayes, decision trees, ensemble methods, supervised neural network models, and more\n",
    "2.  **Unsupervised learning**\n",
    "    *   Model classes include: Clustering, signal decomposition, density estimation, unsupervised neural network models, and more\n",
    "3.  **Model selection and evaluation**\n",
    "    *   Cross-validation, hyper-parameter tuning, model evaluation, model persistence, and validation curves\n",
    "4.  **Dataset transformations**\n",
    "    *   Helpful tools that include: feature extraction, data preprocessing, imputation of missing values, and more\n",
    "5.  **Dataset loading utilities**\n",
    "    *   Toy and example datasets for practice\n",
    "6.  **Computing with scikit-learn**\n",
    "    *   Tools and strategies for optimizing and scaling your computational work"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Where can I find help with using scikit-learn?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The official documentation is a great place to start. There are several links in the documentation that you can consult:\n",
    "\n",
    "*   Official user guide: http://scikit-learn.org/stable/user_guide.html\n",
    "\n",
    "*   Official tutorials: http://scikit-learn.org/stable/tutorial/index.html\n",
    "\n",
    "*   Official examples: http://scikit-learn.org/stable/auto_examples/index.html\n",
    "\n",
    "*   External resources, videos and talks: http://scikit-learn.org/stable/presentations.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### There's a lot there and it's overwhelming, anything to start with first?\n",
    "\n",
    "Give the tutorial titled **A tutorial on statistical-learning for scientific data processing** a look: <http://scikit-learn.org/stable/tutorial/statistical_inference/index.html#stat-learn-tut-index>.\n",
    "\n",
    "Jake VanderPlas puts out a lot of useful tutorials and materials on scientific computing with Python, so the following are worth a look:\n",
    "\n",
    "*   **Tutorial: scikit-learn - Machine Learning in Python with Contributor Jake VanderPlas**: https://youtu.be/cHZONQ2-x7I\n",
    "\n",
    "*   **Scikit-learn tutorials for the Scipy 2013 conference by Gaël Varoquaux, Jake Vanderplas, and Olivier Grisel**\n",
    "\n",
    "    *   *Videos*: Part 1, http://conference.scipy.org/scipy2013/tutorial_detail.php?id=107, and Part 2, http://conference.scipy.org/scipy2013/tutorial_detail.php?id=111\n",
    "\n",
    "    *   *Notebooks for tutorial on GitHub*: https://github.com/jakevdp/sklearn_scipy2013"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### scikit-learn flow chart\n",
    "\n",
    "![Flow chart for scikit-learn](../../img/scikit-learn-flow-chart.png)\n",
    "\n",
    "Interactive version of this chart available here: http://scikit-learn.org/stable/tutorial/machine_learning_map/index.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## scikit-learn worked example: linear regression on Filip dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To ease us into the world of `scikit-learn`, we start by reproducing what we did last time, which was using ordinary least-squares regression to fit the *Filip* dataset using the formula $y=\\beta_{0}+\\beta_{1}x+\\beta_{2}x^{2}+\\beta_{3}x^{3}$. First, in order to use `scikit-learn`, you need to have it installed. The `environment.yaml` file I distributed earlier in the semester should have taken care of this for you. But, just in case it hasn't, it is available through Anaconda with the command:\n",
    "\n",
    "```bash\n",
    "conda install scikit-learn\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, at the top of your notebook, you will need to import the following:\n",
    "\n",
    "```python\n",
    "from sklearn.linear_model import LinearRegression\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unlike with `statsmodels`, where we could write down a formula, in `scikit-learn` we instead need to create columns that represent the different powers in the polynomial. First, let's make a copy of the *Filip* data frame so that we can always go back to the beginning, if necessary:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "filip_df_poly = filip_df.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to create the columns. For the above fit, we need columns that represent $x^2$ and $x^3$. Let's set up a loop to automate creating them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "for n in np.arange(2, 4):\n",
    "    filip_df_poly[f\"x**{n}\"] = filip_df_poly[\"x\"] ** n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the first few rows in the data frame lets us confirm the above loop worked as expected:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y</th>\n",
       "      <th>x</th>\n",
       "      <th>x**2</th>\n",
       "      <th>x**3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.8116</td>\n",
       "      <td>-6.860121</td>\n",
       "      <td>47.061259</td>\n",
       "      <td>-322.845927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.9072</td>\n",
       "      <td>-4.324130</td>\n",
       "      <td>18.698101</td>\n",
       "      <td>-80.853019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.9052</td>\n",
       "      <td>-4.358625</td>\n",
       "      <td>18.997612</td>\n",
       "      <td>-82.803469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.9039</td>\n",
       "      <td>-4.358427</td>\n",
       "      <td>18.995884</td>\n",
       "      <td>-82.792168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.8053</td>\n",
       "      <td>-6.955852</td>\n",
       "      <td>48.383882</td>\n",
       "      <td>-336.551143</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        y         x       x**2        x**3\n",
       "0  0.8116 -6.860121  47.061259 -322.845927\n",
       "1  0.9072 -4.324130  18.698101  -80.853019\n",
       "2  0.9052 -4.358625  18.997612  -82.803469\n",
       "3  0.9039 -4.358427  18.995884  -82.792168\n",
       "4  0.8053 -6.955852  48.383882 -336.551143"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filip_df_poly.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To perform linear regression, we instantiate the `LinearRegression` object as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "lm = LinearRegression()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To create a fit, we use the `lm.fit()` method, which requires two keywords, `X` and `y`. `X` takes the dependent variables we are fitting to (the powers of $x$, in this example). `y` takes the independent variable, or outputs, we are trying to predict with the regression (the variable $y$ in this example). We'll specify the independent variable and dependent variables as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "independent_var = \"y\"\n",
    "dependent_vars = [\"x\", \"x**2\", \"x**3\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now use these in `lm.fit()` as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "smf_filip_fit = lm.fit(X=filip_df_poly[dependent_vars], y=filip_df_poly[independent_var])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The fitting coefficients (the parameters $\\beta_{0}$, $\\beta_{1}$, $\\beta_{2}$, and $\\beta_{3}$) are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>coefficient</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>beta_0</td>\n",
       "      <td>0.390271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>beta_1</td>\n",
       "      <td>-0.303364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>beta_2</td>\n",
       "      <td>-0.053719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>beta_3</td>\n",
       "      <td>-0.002726</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  coefficient     value\n",
       "0      beta_0  0.390271\n",
       "1      beta_1 -0.303364\n",
       "2      beta_2 -0.053719\n",
       "3      beta_3 -0.002726"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame({\n",
    "    \"coefficient\": [\"beta_0\", \"beta_1\", \"beta_2\", \"beta_3\"],\n",
    "    \"value\": np.concatenate([[smf_filip_fit.intercept_], smf_filip_fit.coef_]),\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use the `predict` method to generate our model curve that we can compare with the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "smf_filip_predict_df = pd.DataFrame({\n",
    "    \"x\": np.arange(-9, -3, 0.01),\n",
    "    \"x**2\": np.arange(-9, -3, 0.01) ** 2,\n",
    "    \"x**3\": np.arange(-9, -3, 0.01) ** 3,\n",
    "})\n",
    "smf_filip_predict_df[\"y\"] = smf_filip_fit.predict(smf_filip_predict_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualizing both on the same plot, we get:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(dpi=90)\n",
    "sns.scatterplot(x=\"x\", y=\"y\", data=filip_df, ax=ax)\n",
    "sns.lineplot(x=\"x\", y=\"y\", data=smf_filip_predict_df, ax=ax);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparing with the statsmodels fit, we see that we have the same result:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Statmodels cubic fit to Filip dataset](../../img/filip-statsmodels-cubic-fit.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trying other fits to *Filip*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We've seen all this already. Let's take a look at how the fit to the *Filip* dataset changes as we increase the number of terms in the polynomial expansion. To do this, we'll first need to add even more columns to the data frame so that we can perform a range of fits:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "for n in np.arange(4, 16):\n",
    "    filip_df_poly[f\"x**{n}\"] = filip_df_poly[\"x\"] ** n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have far more columns in the data frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y</th>\n",
       "      <th>x</th>\n",
       "      <th>x**2</th>\n",
       "      <th>x**3</th>\n",
       "      <th>x**4</th>\n",
       "      <th>x**5</th>\n",
       "      <th>x**6</th>\n",
       "      <th>x**7</th>\n",
       "      <th>x**8</th>\n",
       "      <th>x**9</th>\n",
       "      <th>x**10</th>\n",
       "      <th>x**11</th>\n",
       "      <th>x**12</th>\n",
       "      <th>x**13</th>\n",
       "      <th>x**14</th>\n",
       "      <th>x**15</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.8116</td>\n",
       "      <td>-6.860121</td>\n",
       "      <td>47.061259</td>\n",
       "      <td>-322.845927</td>\n",
       "      <td>2214.762094</td>\n",
       "      <td>-15193.535763</td>\n",
       "      <td>104229.492448</td>\n",
       "      <td>-715026.920996</td>\n",
       "      <td>4.905171e+06</td>\n",
       "      <td>-3.365007e+07</td>\n",
       "      <td>2.308435e+08</td>\n",
       "      <td>-1.583615e+09</td>\n",
       "      <td>1.086379e+10</td>\n",
       "      <td>-7.452689e+10</td>\n",
       "      <td>5.112635e+11</td>\n",
       "      <td>-3.507329e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.9072</td>\n",
       "      <td>-4.324130</td>\n",
       "      <td>18.698101</td>\n",
       "      <td>-80.853019</td>\n",
       "      <td>349.618968</td>\n",
       "      <td>-1511.797883</td>\n",
       "      <td>6537.210647</td>\n",
       "      <td>-28267.748970</td>\n",
       "      <td>1.222334e+05</td>\n",
       "      <td>-5.285532e+05</td>\n",
       "      <td>2.285533e+06</td>\n",
       "      <td>-9.882941e+06</td>\n",
       "      <td>4.273512e+07</td>\n",
       "      <td>-1.847922e+08</td>\n",
       "      <td>7.990656e+08</td>\n",
       "      <td>-3.455264e+09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.9052</td>\n",
       "      <td>-4.358625</td>\n",
       "      <td>18.997612</td>\n",
       "      <td>-82.803469</td>\n",
       "      <td>360.909276</td>\n",
       "      <td>-1573.068212</td>\n",
       "      <td>6856.414522</td>\n",
       "      <td>-29884.540122</td>\n",
       "      <td>1.302555e+05</td>\n",
       "      <td>-5.677349e+05</td>\n",
       "      <td>2.474544e+06</td>\n",
       "      <td>-1.078561e+07</td>\n",
       "      <td>4.701042e+07</td>\n",
       "      <td>-2.049008e+08</td>\n",
       "      <td>8.930857e+08</td>\n",
       "      <td>-3.892626e+09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.9039</td>\n",
       "      <td>-4.358427</td>\n",
       "      <td>18.995884</td>\n",
       "      <td>-82.792168</td>\n",
       "      <td>360.843598</td>\n",
       "      <td>-1572.710389</td>\n",
       "      <td>6854.543023</td>\n",
       "      <td>-29875.023648</td>\n",
       "      <td>1.302081e+05</td>\n",
       "      <td>-5.675025e+05</td>\n",
       "      <td>2.473418e+06</td>\n",
       "      <td>-1.078021e+07</td>\n",
       "      <td>4.698476e+07</td>\n",
       "      <td>-2.047796e+08</td>\n",
       "      <td>8.925170e+08</td>\n",
       "      <td>-3.889970e+09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.8053</td>\n",
       "      <td>-6.955852</td>\n",
       "      <td>48.383882</td>\n",
       "      <td>-336.551143</td>\n",
       "      <td>2341.000068</td>\n",
       "      <td>-16283.650894</td>\n",
       "      <td>113266.671807</td>\n",
       "      <td>-787866.248553</td>\n",
       "      <td>5.480281e+06</td>\n",
       "      <td>-3.812003e+07</td>\n",
       "      <td>2.651573e+08</td>\n",
       "      <td>-1.844395e+09</td>\n",
       "      <td>1.282934e+10</td>\n",
       "      <td>-8.923899e+10</td>\n",
       "      <td>6.207332e+11</td>\n",
       "      <td>-4.317729e+12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        y         x       x**2        x**3         x**4          x**5  \\\n",
       "0  0.8116 -6.860121  47.061259 -322.845927  2214.762094 -15193.535763   \n",
       "1  0.9072 -4.324130  18.698101  -80.853019   349.618968  -1511.797883   \n",
       "2  0.9052 -4.358625  18.997612  -82.803469   360.909276  -1573.068212   \n",
       "3  0.9039 -4.358427  18.995884  -82.792168   360.843598  -1572.710389   \n",
       "4  0.8053 -6.955852  48.383882 -336.551143  2341.000068 -16283.650894   \n",
       "\n",
       "            x**6           x**7          x**8          x**9         x**10  \\\n",
       "0  104229.492448 -715026.920996  4.905171e+06 -3.365007e+07  2.308435e+08   \n",
       "1    6537.210647  -28267.748970  1.222334e+05 -5.285532e+05  2.285533e+06   \n",
       "2    6856.414522  -29884.540122  1.302555e+05 -5.677349e+05  2.474544e+06   \n",
       "3    6854.543023  -29875.023648  1.302081e+05 -5.675025e+05  2.473418e+06   \n",
       "4  113266.671807 -787866.248553  5.480281e+06 -3.812003e+07  2.651573e+08   \n",
       "\n",
       "          x**11         x**12         x**13         x**14         x**15  \n",
       "0 -1.583615e+09  1.086379e+10 -7.452689e+10  5.112635e+11 -3.507329e+12  \n",
       "1 -9.882941e+06  4.273512e+07 -1.847922e+08  7.990656e+08 -3.455264e+09  \n",
       "2 -1.078561e+07  4.701042e+07 -2.049008e+08  8.930857e+08 -3.892626e+09  \n",
       "3 -1.078021e+07  4.698476e+07 -2.047796e+08  8.925170e+08 -3.889970e+09  \n",
       "4 -1.844395e+09  1.282934e+10 -8.923899e+10  6.207332e+11 -4.317729e+12  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filip_df_poly.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We set up a loop to create a series of fits:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "independent_var = \"y\"\n",
    "dependent_vars = [\"x\"]\n",
    "\n",
    "filip_polyfit_predict_df = pd.DataFrame({\n",
    "    \"x\": np.arange(-8.90, -2.9, 0.01),\n",
    "})\n",
    "\n",
    "for n in range(1, 16):\n",
    "    if n > 1:\n",
    "        dependent_vars.append(f\"x**{n}\")\n",
    "        filip_polyfit_predict_df[f\"x**{n}\"] = \\\n",
    "            np.arange(-8.90, -2.9, 0.01) ** n\n",
    "        \n",
    "    filip_polyfit = lm.fit(\n",
    "        X=filip_df_poly[dependent_vars],\n",
    "        y=filip_df_poly[independent_var],\n",
    "    )\n",
    "    filip_polyfit_predict_df[f\"{n}\"] = \\\n",
    "        filip_polyfit.predict(filip_polyfit_predict_df[dependent_vars])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's reshape the data frame using melt. Instead of having 15 columns named `y_1` through `y_15`, we'll instead have a column called `n` that specifies the polynomial degree and a column `y_predicted` that contains the predicted value of $y$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "filip_polyfit_predict_df_reshaped = filip_polyfit_predict_df \\\n",
    "    .melt(id_vars=dependent_vars,\n",
    "          value_vars=[f\"{n}\" for n in range(1, 16)],\n",
    "          var_name=\"Degree\",\n",
    "          value_name=\"y_predicted\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now view the results in a series of faceted windows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x900 with 15 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(\n",
    "    filip_polyfit_predict_df_reshaped,\n",
    "    col=\"Degree\",\n",
    "    col_wrap=4,\n",
    "    col_order=[f\"{n}\" for n in range(1, 16)],\n",
    "    hue_kws={\"color\": [\"k\"]},\n",
    "    sharex=True,\n",
    "    sharey=True,\n",
    ")\n",
    "\n",
    "# Plots the predicted values as black lines\n",
    "g = g.map(plt.plot, \"x\", \"y_predicted\")\n",
    "\n",
    "# Plots the data points within each facet\n",
    "for idx in range(15):\n",
    "    sns.scatterplot(x=\"x\", y=\"y\", data=filip_df_poly, ax=g.axes[idx])\n",
    "\n",
    "g.fig.set_dpi(150)\n",
    "g.fig.set_size_inches(6, 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An alternate way to plot the windows, which is a little more complicated, but offers more control, is as follows:\n",
    "\n",
    "```python\n",
    "fig, ax = plt.subplots(nrows=4, ncols=4, figsize=(8, 8), dpi=120,\n",
    "                       sharex=True, sharey=True)\n",
    "\n",
    "poly_degree = 0\n",
    "for nrow in range(4):\n",
    "    for ncol in range(4):\n",
    "        poly_degree += 1\n",
    "        \n",
    "        if poly_degree < 16:\n",
    "            sns.scatterplot(\n",
    "                x=\"x\",\n",
    "                y=\"y\",\n",
    "                data=filip_df_poly, ax=ax[nrow][ncol]\n",
    "            )\n",
    "            sns.lineplot(\n",
    "                x=\"x\",\n",
    "                y=f\"y_{poly_degree}\",\n",
    "                data=smf_filip_polyfit_predict_df,\n",
    "                color=\"black\",\n",
    "                ax=ax[nrow][ncol]\n",
    "            )\n",
    "            ax[nrow][ncol].set_title(f\"Degree {poly_degree}\")\n",
    "            ax[nrow][ncol].set_ylabel(\"y\")\n",
    "            \n",
    "plt.tight_layout();\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**So... which model is best? How can we even tell?**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model selection using cross-validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* How should we compare and rank models?\n",
    "\n",
    "* This is what **model selection** is about, computing scores and measures of model performance for different models, and selecting the best choice.\n",
    "\n",
    "* Cross-validation is a method that can be used to compare relative model performance using only training data \n",
    "\n",
    "* Cross-validation refers to the procedure of removing portions of a dataset, training the model on the remaining portion, and then trying to predict the values of the data you removed\n",
    "\n",
    "* A popular flavor of cross-validation (especially among data scientists) is called **k-fold cross-validation**\n",
    "\n",
    "* **Basic idea:** Estimate how robust your model is by systematically removing different chunks (the \"folds\") of the dataset, repeating the fitting process, then testing its predictive power on the folds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![cross validation schematic](../../img/cross-validation-schematic.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* The above example illustrates a *5-fold*, or $k=5$, cross-validation.\n",
    "\n",
    "* Each fold will act as a testing set, with the remaining $k-1$ folds used to train the model.\n",
    "\n",
    "* Fit model, predict values in testing set, then calculate a score, such as the mean-squared prediction error (MSE)\n",
    "\n",
    "* MSE gives an estimate of how well the model works as a predictor\n",
    "\n",
    "* MSE is general-purpose and allows you to compare models of many types\n",
    "\n",
    "* For linear regression, you can also use model-specific measures, such as $R^2$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross-validation using `scikit-learn`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To use cross-validation, you will need to import additional modules from `scikit-learn`:\n",
    "\n",
    "```python\n",
    "from sklearn.model_selection import cross_val_score, RepeatedKFold\n",
    "```\n",
    "\n",
    "`RepeatedKFold` is an implementation of *k-fold cross-validation* that allows you to repeat the procedure multiple times, randomizing the data within each repetition. By doing this, you can obtain better statistical averaging, which can be particularly useful if the dataset is not so large.\n",
    "\n",
    "`cross_val_score` is the function that takes the data and cross-validation splitting as input, runs the fitting procedure, and computes a specified score at the end."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A list of scores you can compute during a cross-validation run:** <http://scikit-learn.org/stable/modules/model_evaluation.html#the-scoring-parameter-defining-model-evaluation-rules>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's modify our loop to run a series of k-fold cross-validations for our different polynomial formulas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "independent_var = \"y\"\n",
    "dependent_vars = [\"x\"]\n",
    "\n",
    "filip_poly_cv_scores = {\n",
    "    \"n\": [],\n",
    "    \"mse\": [],\n",
    "    \"mse_sd\": [],\n",
    "    \"r**2\": [],\n",
    "    \"r**2_sd\": [],\n",
    "}\n",
    "\n",
    "for n in range(1, 16):\n",
    "    if n > 1:\n",
    "        dependent_vars.append(f\"x**{n}\")\n",
    "        \n",
    "    rkf = RepeatedKFold(\n",
    "        n_splits=10,\n",
    "        n_repeats=100,\n",
    "        random_state=int(np.round(np.random.uniform(0, 2**31), decimals=0)),\n",
    "    )\n",
    "    \n",
    "    # Cross-validated mean-squared error score\n",
    "    mse_cv_score = cross_val_score(\n",
    "        lm,\n",
    "        filip_df_poly[dependent_vars],\n",
    "        filip_df_poly[independent_var],\n",
    "        scoring=\"neg_mean_squared_error\",\n",
    "        cv=rkf,\n",
    "        n_jobs=-1,  # Use all processors during cross-validation run\n",
    "    )\n",
    "\n",
    "    # Cross-validated R**2 score\n",
    "    r2_cv_score = cross_val_score(\n",
    "        lm,\n",
    "        filip_df_poly[dependent_vars],\n",
    "        filip_df_poly[independent_var],\n",
    "        scoring=\"r2\",\n",
    "        cv=rkf,\n",
    "        n_jobs=-1,  # Use all processors during cross-validation run\n",
    "    )\n",
    "\n",
    "    filip_poly_cv_scores[\"n\"].append(n)\n",
    "    filip_poly_cv_scores[\"mse\"].append(-np.mean(mse_cv_score))  # Get rid of negative sign\n",
    "    filip_poly_cv_scores[\"mse_sd\"].append(np.std(mse_cv_score))\n",
    "    filip_poly_cv_scores[\"r**2\"].append(np.mean(r2_cv_score))\n",
    "    filip_poly_cv_scores[\"r**2_sd\"].append(np.std(r2_cv_score))\n",
    "\n",
    "# Convert dictionary to data frame\n",
    "filip_poly_cv_scores_df = pd.DataFrame(filip_poly_cv_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how the various models perform:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n</th>\n",
       "      <th>mse</th>\n",
       "      <th>mse_sd</th>\n",
       "      <th>r**2</th>\n",
       "      <th>r**2_sd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.000387</td>\n",
       "      <td>0.000120</td>\n",
       "      <td>0.815206</td>\n",
       "      <td>0.245704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.000299</td>\n",
       "      <td>0.000115</td>\n",
       "      <td>0.860383</td>\n",
       "      <td>0.147108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0.000221</td>\n",
       "      <td>0.000082</td>\n",
       "      <td>0.887993</td>\n",
       "      <td>0.210553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.000093</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>0.959409</td>\n",
       "      <td>0.033963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.000102</td>\n",
       "      <td>0.000071</td>\n",
       "      <td>0.955175</td>\n",
       "      <td>0.044956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0.000040</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>0.981716</td>\n",
       "      <td>0.022905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>0.000062</td>\n",
       "      <td>0.975425</td>\n",
       "      <td>0.132341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>0.000078</td>\n",
       "      <td>0.984792</td>\n",
       "      <td>0.036522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>0.000030</td>\n",
       "      <td>0.000126</td>\n",
       "      <td>0.986491</td>\n",
       "      <td>0.061791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>0.000098</td>\n",
       "      <td>0.988350</td>\n",
       "      <td>0.038857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>0.992845</td>\n",
       "      <td>0.011417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.994672</td>\n",
       "      <td>0.005599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>13</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.994456</td>\n",
       "      <td>0.007132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>0.992566</td>\n",
       "      <td>0.032214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.993008</td>\n",
       "      <td>0.009973</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     n       mse    mse_sd      r**2   r**2_sd\n",
       "0    1  0.000387  0.000120  0.815206  0.245704\n",
       "1    2  0.000299  0.000115  0.860383  0.147108\n",
       "2    3  0.000221  0.000082  0.887993  0.210553\n",
       "3    4  0.000093  0.000035  0.959409  0.033963\n",
       "4    5  0.000102  0.000071  0.955175  0.044956\n",
       "5    6  0.000040  0.000025  0.981716  0.022905\n",
       "6    7  0.000044  0.000062  0.975425  0.132341\n",
       "7    8  0.000035  0.000078  0.984792  0.036522\n",
       "8    9  0.000030  0.000126  0.986491  0.061791\n",
       "9   10  0.000028  0.000098  0.988350  0.038857\n",
       "10  11  0.000016  0.000022  0.992845  0.011417\n",
       "11  12  0.000012  0.000008  0.994672  0.005599\n",
       "12  13  0.000012  0.000012  0.994456  0.007132\n",
       "13  14  0.000016  0.000027  0.992566  0.032214\n",
       "14  15  0.000016  0.000024  0.993008  0.009973"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filip_poly_cv_scores_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize the relative performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 480x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(nrows=2, dpi=120, figsize=(4, 6), sharex=True)\n",
    "sns.scatterplot(x=\"n\", y=\"mse\", data=filip_poly_cv_scores_df, ax=ax[0])\n",
    "sns.lineplot(x=\"n\", y=\"mse\", data=filip_poly_cv_scores_df, ax=ax[0])\n",
    "ax[0].axvline(x=10, linestyle=\"--\", color=\"k\")  # \"True\" polynomial degree\n",
    "ax[0].set_ylim([-0.0001, 0.0005])\n",
    "ax[0].set_xlabel(r\"Polynomial degree\")\n",
    "ax[0].set_ylabel(r\"Mean-squared error\")\n",
    "\n",
    "sns.scatterplot(x=\"n\", y=\"r**2\", data=filip_poly_cv_scores_df, ax=ax[1])\n",
    "sns.lineplot(x=\"n\", y=\"r**2\", data=filip_poly_cv_scores_df, ax=ax[1])\n",
    "ax[1].axvline(x=10, linestyle=\"--\", color=\"k\")  # \"True\" polynomial degree\n",
    "ax[1].set_ylim([0.80, 1.05])\n",
    "ax[1].set_xlabel(r\"Polynomial degree\")\n",
    "ax[1].set_ylabel(r\"$R^{2}$\")\n",
    "\n",
    "fig.tight_layout();"
   ]
  }
 ],
 "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": "3.6.5"
  },
  "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()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}