ml-finance-python

python scripts for finance machine learning

git clone https://9o.is/git/ml-finance-python.git

02_random_forest.ipynb

(379978B)


      1 {
      2  "cells": [
      3   {
      4    "cell_type": "markdown",
      5    "metadata": {},
      6    "source": [
      7     "# How to train and tune a random forest"
      8    ]
      9   },
     10   {
     11    "cell_type": "code",
     12    "execution_count": 4,
     13    "metadata": {
     14     "ExecuteTime": {
     15      "end_time": "2018-11-09T16:21:43.085379Z",
     16      "start_time": "2018-11-09T16:21:43.081044Z"
     17     }
     18    },
     19    "outputs": [],
     20    "source": [
     21     "%matplotlib inline\n",
     22     "\n",
     23     "import warnings\n",
     24     "import os\n",
     25     "from pathlib import Path\n",
     26     "import quandl\n",
     27     "import numpy as np\n",
     28     "from numpy.random import choice, normal\n",
     29     "import pandas as pd\n",
     30     "\n",
     31     "import matplotlib.pyplot as plt\n",
     32     "import seaborn as sns\n",
     33     "\n",
     34     "from sklearn.tree import DecisionTreeRegressor, DecisionTreeClassifier\n",
     35     "from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier, BaggingClassifier, BaggingRegressor\n",
     36     "from sklearn.model_selection import train_test_split, cross_val_score, GridSearchCV, cross_val_score\n",
     37     "from sklearn.metrics import roc_auc_score, roc_curve, mean_squared_error, precision_recall_curve\n",
     38     "from sklearn.externals import joblib\n",
     39     "warnings.filterwarnings('ignore')"
     40    ]
     41   },
     42   {
     43    "cell_type": "code",
     44    "execution_count": 5,
     45    "metadata": {
     46     "ExecuteTime": {
     47      "end_time": "2018-11-02T22:34:06.704816Z",
     48      "start_time": "2018-11-02T22:34:06.697905Z"
     49     }
     50    },
     51    "outputs": [],
     52    "source": [
     53     "plt.style.use('fivethirtyeight')\n",
     54     "np.random.seed(seed=42)"
     55    ]
     56   },
     57   {
     58    "cell_type": "markdown",
     59    "metadata": {},
     60    "source": [
     61     "## Get Data"
     62    ]
     63   },
     64   {
     65    "cell_type": "code",
     66    "execution_count": 35,
     67    "metadata": {
     68     "ExecuteTime": {
     69      "end_time": "2018-11-09T16:08:12.421063Z",
     70      "start_time": "2018-11-09T16:08:12.223895Z"
     71     }
     72    },
     73    "outputs": [
     74     {
     75      "name": "stdout",
     76      "output_type": "stream",
     77      "text": [
     78       "<class 'pandas.core.frame.DataFrame'>\n",
     79       "MultiIndex: 171162 entries, (A, 2011-03-31 00:00:00) to (ZUMZ, 2018-02-28 00:00:00)\n",
     80       "Data columns (total 61 columns):\n",
     81       "returns                  171162 non-null float64\n",
     82       "t-1                      171162 non-null float64\n",
     83       "t-2                      171162 non-null float64\n",
     84       "t-3                      171162 non-null float64\n",
     85       "t-4                      171162 non-null float64\n",
     86       "t-5                      171162 non-null float64\n",
     87       "t-6                      171162 non-null float64\n",
     88       "t-7                      171162 non-null float64\n",
     89       "t-8                      171162 non-null float64\n",
     90       "t-9                      171162 non-null float64\n",
     91       "t-10                     171162 non-null float64\n",
     92       "t-11                     171162 non-null float64\n",
     93       "t-12                     171162 non-null float64\n",
     94       "year_2011                171162 non-null uint8\n",
     95       "year_2012                171162 non-null uint8\n",
     96       "year_2013                171162 non-null uint8\n",
     97       "year_2014                171162 non-null uint8\n",
     98       "year_2015                171162 non-null uint8\n",
     99       "year_2016                171162 non-null uint8\n",
    100       "year_2017                171162 non-null uint8\n",
    101       "year_2018                171162 non-null uint8\n",
    102       "month_1                  171162 non-null uint8\n",
    103       "month_2                  171162 non-null uint8\n",
    104       "month_3                  171162 non-null uint8\n",
    105       "month_4                  171162 non-null uint8\n",
    106       "month_5                  171162 non-null uint8\n",
    107       "month_6                  171162 non-null uint8\n",
    108       "month_7                  171162 non-null uint8\n",
    109       "month_8                  171162 non-null uint8\n",
    110       "month_9                  171162 non-null uint8\n",
    111       "month_10                 171162 non-null uint8\n",
    112       "month_11                 171162 non-null uint8\n",
    113       "month_12                 171162 non-null uint8\n",
    114       "size_1                   171162 non-null int8\n",
    115       "size_2                   171162 non-null int8\n",
    116       "size_3                   171162 non-null int8\n",
    117       "size_4                   171162 non-null int8\n",
    118       "size_5                   171162 non-null int8\n",
    119       "size_6                   171162 non-null int8\n",
    120       "size_7                   171162 non-null int8\n",
    121       "size_8                   171162 non-null int8\n",
    122       "size_9                   171162 non-null int8\n",
    123       "size_10                  171162 non-null int8\n",
    124       "age_0                    171162 non-null int8\n",
    125       "age_1                    171162 non-null int8\n",
    126       "age_2                    171162 non-null int8\n",
    127       "age_3                    171162 non-null int8\n",
    128       "age_4                    171162 non-null int8\n",
    129       "age_5                    171162 non-null int8\n",
    130       "Basic Industries         171162 non-null int8\n",
    131       "Capital Goods            171162 non-null int8\n",
    132       "Consumer Durables        171162 non-null int8\n",
    133       "Consumer Non-Durables    171162 non-null int8\n",
    134       "Consumer Services        171162 non-null int8\n",
    135       "Energy                   171162 non-null int8\n",
    136       "Finance                  171162 non-null int8\n",
    137       "Health Care              171162 non-null int8\n",
    138       "Miscellaneous            171162 non-null int8\n",
    139       "Public Utilities         171162 non-null int8\n",
    140       "Technology               171162 non-null int8\n",
    141       "Transportation           171162 non-null int8\n",
    142       "dtypes: float64(13), int8(28), uint8(20)\n",
    143       "memory usage: 25.4+ MB\n"
    144      ]
    145     }
    146    ],
    147    "source": [
    148     "with pd.HDFStore('data.h5') as store:\n",
    149     "    data =store['data']\n",
    150     "data.info()"
    151    ]
    152   },
    153   {
    154    "cell_type": "markdown",
    155    "metadata": {},
    156    "source": [
    157     "### Stock Prices"
    158    ]
    159   },
    160   {
    161    "cell_type": "code",
    162    "execution_count": 36,
    163    "metadata": {
    164     "ExecuteTime": {
    165      "end_time": "2018-11-09T16:08:23.234002Z",
    166      "start_time": "2018-11-09T16:08:23.216783Z"
    167     }
    168    },
    169    "outputs": [],
    170    "source": [
    171     "y = data.returns\n",
    172     "y_binary = (y > 0).astype(int)\n",
    173     "X = data.drop('returns', axis=1)"
    174    ]
    175   },
    176   {
    177    "cell_type": "markdown",
    178    "metadata": {},
    179    "source": [
    180     "## Explore Data"
    181    ]
    182   },
    183   {
    184    "cell_type": "code",
    185    "execution_count": 8,
    186    "metadata": {
    187     "ExecuteTime": {
    188      "end_time": "2018-11-09T16:08:27.274474Z",
    189      "start_time": "2018-11-09T16:08:27.227505Z"
    190     }
    191    },
    192    "outputs": [
    193     {
    194      "data": {
    195       "text/plain": [
    196        "count    171162.000000\n",
    197        "mean          0.009827\n",
    198        "std           0.068340\n",
    199        "min          -0.152427\n",
    200        "10%          -0.080626\n",
    201        "20%          -0.048064\n",
    202        "30.0%        -0.025061\n",
    203        "40%          -0.006481\n",
    204        "50%           0.009259\n",
    205        "60%           0.025641\n",
    206        "70%           0.043847\n",
    207        "80%           0.066246\n",
    208        "90%           0.100111\n",
    209        "max           0.185083\n",
    210        "Name: returns, dtype: float64"
    211       ]
    212      },
    213      "execution_count": 8,
    214      "metadata": {},
    215      "output_type": "execute_result"
    216     }
    217    ],
    218    "source": [
    219     "y.describe(percentiles=np.arange(.1, .91, .1))"
    220    ]
    221   },
    222   {
    223    "cell_type": "markdown",
    224    "metadata": {},
    225    "source": [
    226     "## Custom KFold"
    227    ]
    228   },
    229   {
    230    "cell_type": "code",
    231    "execution_count": 9,
    232    "metadata": {
    233     "ExecuteTime": {
    234      "end_time": "2018-11-09T16:08:43.761130Z",
    235      "start_time": "2018-11-09T16:08:43.737859Z"
    236     }
    237    },
    238    "outputs": [],
    239    "source": [
    240     "class OneStepTimeSeriesSplit:\n",
    241     "    \"\"\"Generates tuples of train_idx, test_idx pairs\n",
    242     "    Assumes the index contains a level labeled 'date'\"\"\"\n",
    243     "\n",
    244     "    def __init__(self, n_splits=3, test_period_length=1, shuffle=False):\n",
    245     "        self.n_splits = n_splits\n",
    246     "        self.test_period_length = test_period_length\n",
    247     "        self.shuffle = shuffle\n",
    248     "\n",
    249     "    @staticmethod\n",
    250     "    def chunks(l, n):\n",
    251     "        for i in range(0, len(l), n):\n",
    252     "            yield l[i:i + n]\n",
    253     "\n",
    254     "    def split(self, X, y=None, groups=None):\n",
    255     "        unique_dates = (X.index\n",
    256     "                        .get_level_values('date')\n",
    257     "                        .unique()\n",
    258     "                        .sort_values(ascending=False)\n",
    259     "                        [:self.n_splits*self.test_period_length])\n",
    260     "\n",
    261     "        dates = X.reset_index()[['date']]\n",
    262     "        for test_date in self.chunks(unique_dates, self.test_period_length):\n",
    263     "            train_idx = dates[dates.date < min(test_date)].index\n",
    264     "            test_idx = dates[dates.date.isin(test_date)].index\n",
    265     "            if self.shuffle:\n",
    266     "                np.random.shuffle(list(train_idx))\n",
    267     "            yield train_idx, test_idx\n",
    268     "\n",
    269     "    def get_n_splits(self, X, y, groups=None):\n",
    270     "        return self.n_splits"
    271    ]
    272   },
    273   {
    274    "cell_type": "markdown",
    275    "metadata": {},
    276    "source": [
    277     "## Benchmarks"
    278    ]
    279   },
    280   {
    281    "cell_type": "markdown",
    282    "metadata": {},
    283    "source": [
    284     "### Regression"
    285    ]
    286   },
    287   {
    288    "cell_type": "code",
    289    "execution_count": 10,
    290    "metadata": {
    291     "ExecuteTime": {
    292      "end_time": "2018-11-09T16:08:44.090957Z",
    293      "start_time": "2018-11-09T16:08:44.085072Z"
    294     }
    295    },
    296    "outputs": [],
    297    "source": [
    298     "def regression_benchmark():\n",
    299     "    rmse = []\n",
    300     "    for train_idx, test_idx in cv.split(X):\n",
    301     "        mean = y.iloc[train_idx].mean()\n",
    302     "        data = y.iloc[test_idx].to_frame('y_test').assign(y_pred=mean)\n",
    303     "        rmse.append(np.sqrt(mean_squared_error(data.y_test, data.y_pred))) \n",
    304     "    return np.mean(rmse)"
    305    ]
    306   },
    307   {
    308    "cell_type": "markdown",
    309    "metadata": {},
    310    "source": [
    311     "### Classification"
    312    ]
    313   },
    314   {
    315    "cell_type": "code",
    316    "execution_count": 11,
    317    "metadata": {
    318     "ExecuteTime": {
    319      "end_time": "2018-11-09T16:08:46.622752Z",
    320      "start_time": "2018-11-09T16:08:46.620513Z"
    321     }
    322    },
    323    "outputs": [],
    324    "source": [
    325     "def classification_benchmark():\n",
    326     "    auc = []\n",
    327     "    for train_idx, test_idx in cv.split(X):\n",
    328     "        mean = y_binary.iloc[train_idx].mean()\n",
    329     "        data = y_binary.iloc[test_idx].to_frame('y_test').assign(y_pred=mean)\n",
    330     "        auc.append(roc_auc_score(data.y_test, data.y_pred))\n",
    331     "    return np.mean(auc)"
    332    ]
    333   },
    334   {
    335    "cell_type": "markdown",
    336    "metadata": {},
    337    "source": [
    338     "## Bagged Decision Trees"
    339    ]
    340   },
    341   {
    342    "cell_type": "markdown",
    343    "metadata": {},
    344    "source": [
    345     "To apply bagging to decision trees, we create bootstrap samples from our training data by repeatedly sampling with replacement, then train one decision tree on each of these samples, and create an ensemble prediction by averaging over the predictions of the different trees.\n",
    346     "\n",
    347     "Bagged decision trees are usually grown large, that is, have many levels and leaf nodes and are not pruned so that each tree has low bias but high variance. The effect of averaging their predictions then aims to reduce their variance. Bagging has been shown to substantially improve predictive performance by constructing ensembles that combine hundreds or even thousands of trees trained on bootstrap samples.\n",
    348     "\n",
    349     "To illustrate the effect of bagging on the variance of a regression tree, we can use the `BaggingRegressor` meta-estimator provided by `sklearn`. It trains a user-defined base estimator based on parameters that specify the sampling strategy:"
    350    ]
    351   },
    352   {
    353    "cell_type": "markdown",
    354    "metadata": {},
    355    "source": [
    356     "- `max_samples` and `max_features` control the size of the subsets drawn from the rows and the columns, respectively\n",
    357     "- `bootstrap` and `bootstrap_features` determine whether each of these samples is drawn with or without replacement"
    358    ]
    359   },
    360   {
    361    "cell_type": "code",
    362    "execution_count": 12,
    363    "metadata": {
    364     "ExecuteTime": {
    365      "end_time": "2018-11-09T16:08:47.098413Z",
    366      "start_time": "2018-11-09T16:08:46.924454Z"
    367     },
    368     "scrolled": true
    369    },
    370    "outputs": [
    371     {
    372      "data": {
    373       "image/png": "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\n",
    374       "text/plain": [
    375        "<Figure size 432x288 with 1 Axes>"
    376       ]
    377      },
    378      "metadata": {},
    379      "output_type": "display_data"
    380     }
    381    ],
    382    "source": [
    383     "def f(x):\n",
    384     "    return np.exp(-(x+2) ** 2) + np.cos((x-2)**2)\n",
    385     "\n",
    386     "x = np.linspace(-3, 2, 1000)\n",
    387     "y = pd.Series(f(x), index=x)\n",
    388     "y.plot();"
    389    ]
    390   },
    391   {
    392    "cell_type": "markdown",
    393    "metadata": {},
    394    "source": [
    395     "The following example uses the preceding exponential function `f(x)` to generate training samples for a single DecisionTreeRegressor and a BaggingRegressor ensemble that consists of ten trees, each grown ten levels deep. Both models are trained on the random samples and predict outcomes for the actual function with added noise.\n",
    396     "\n",
    397     "Since we know the true function, we can decompose the mean-squared error into bias, variance, and noise, and compare the relative size of these components for both models according to the following breakdown:\n",
    398     "\n",
    399     "For 100 repeated random training and test samples of 250 and 500 observations each, we find that the variance of the predictions of the individual decision tree is almost twice as high as that for the small ensemble of 10 bagged trees based on bootstrapped samples:"
    400    ]
    401   },
    402   {
    403    "cell_type": "code",
    404    "execution_count": 13,
    405    "metadata": {
    406     "ExecuteTime": {
    407      "end_time": "2018-11-09T16:08:47.818504Z",
    408      "start_time": "2018-11-09T16:08:47.099798Z"
    409     }
    410    },
    411    "outputs": [],
    412    "source": [
    413     "test_size = 500\n",
    414     "train_size = 250\n",
    415     "reps = 100\n",
    416     "\n",
    417     "noise = .5  # noise relative to std(y)\n",
    418     "noise = y.std() * noise\n",
    419     "\n",
    420     "X_test = choice(x, size=test_size, replace=False)\n",
    421     "\n",
    422     "max_depth = 10\n",
    423     "n_estimators=10\n",
    424     "\n",
    425     "tree = DecisionTreeRegressor(max_depth=max_depth)\n",
    426     "bagged_tree = BaggingRegressor(base_estimator=tree, n_estimators=n_estimators)\n",
    427     "learners = {'Decision Tree': tree, 'Bagging Regressor': bagged_tree}\n",
    428     "\n",
    429     "predictions = {k: pd.DataFrame() for k, v in learners.items()}\n",
    430     "for i in range(reps):\n",
    431     "    X_train = choice(x, train_size)\n",
    432     "    y_train = f(X_train) + normal(scale=noise, size=train_size)\n",
    433     "    for label, learner in learners.items():\n",
    434     "        learner.fit(X=X_train.reshape(-1, 1), y=y_train)\n",
    435     "        preds = pd.DataFrame({i: learner.predict(X_test.reshape(-1, 1))}, index=X_test)\n",
    436     "        predictions[label] = pd.concat([predictions[label], preds], axis=1)"
    437    ]
    438   },
    439   {
    440    "cell_type": "code",
    441    "execution_count": 14,
    442    "metadata": {
    443     "ExecuteTime": {
    444      "end_time": "2018-11-09T16:08:47.823493Z",
    445      "start_time": "2018-11-09T16:08:47.819814Z"
    446     }
    447    },
    448    "outputs": [],
    449    "source": [
    450     "# y only observed with noise\n",
    451     "y_true = pd.Series(f(X_test), index=X_test)\n",
    452     "y_test = pd.DataFrame(y_true.values.reshape(-1,1) + normal(scale=noise, size=(test_size, reps)), index=X_test)"
    453    ]
    454   },
    455   {
    456    "cell_type": "code",
    457    "execution_count": 15,
    458    "metadata": {
    459     "ExecuteTime": {
    460      "end_time": "2018-11-09T16:08:50.109678Z",
    461      "start_time": "2018-11-09T16:08:50.084044Z"
    462     }
    463    },
    464    "outputs": [],
    465    "source": [
    466     "result = pd.DataFrame()\n",
    467     "for label, preds in predictions.items():\n",
    468     "    result[(label, 'error')] = preds.sub(y_test, axis=0).pow(2).mean(1)    # mean squared error\n",
    469     "    result[(label, 'bias')] = y_true.sub(preds.mean(axis=1), axis=0).pow(2)             # bias\n",
    470     "    result[(label, 'variance')] = preds.var(axis=1)\n",
    471     "    result[(label, 'noise', )] = y_test.var(axis=1)\n",
    472     "result.columns = pd.MultiIndex.from_tuples(result.columns)"
    473    ]
    474   },
    475   {
    476    "cell_type": "code",
    477    "execution_count": 16,
    478    "metadata": {
    479     "ExecuteTime": {
    480      "end_time": "2018-11-09T16:08:50.246658Z",
    481      "start_time": "2018-11-09T16:08:50.236695Z"
    482     }
    483    },
    484    "outputs": [
    485     {
    486      "data": {
    487       "text/plain": [
    488        "'0.29%'"
    489       ]
    490      },
    491      "execution_count": 16,
    492      "metadata": {},
    493      "output_type": "execute_result"
    494     }
    495    ],
    496    "source": [
    497     "df = result.mean().sort_index().loc['Decision Tree']\n",
    498     "f\"{(df.error- df.drop('error').sum()) / df.error:.2%}\""
    499    ]
    500   },
    501   {
    502    "cell_type": "code",
    503    "execution_count": 17,
    504    "metadata": {
    505     "ExecuteTime": {
    506      "end_time": "2018-11-09T16:08:50.355402Z",
    507      "start_time": "2018-11-09T16:08:50.349303Z"
    508     }
    509    },
    510    "outputs": [
    511     {
    512      "data": {
    513       "text/plain": [
    514        "'0.22%'"
    515       ]
    516      },
    517      "execution_count": 17,
    518      "metadata": {},
    519      "output_type": "execute_result"
    520     }
    521    ],
    522    "source": [
    523     "df = result.mean().sort_index().loc['Bagging Regressor']\n",
    524     "f\"{(df.error- df.drop('error').sum()) / df.error:.2%}\""
    525    ]
    526   },
    527   {
    528    "cell_type": "markdown",
    529    "metadata": {},
    530    "source": [
    531     "### Visualize Bias-Variance Breakdown"
    532    ]
    533   },
    534   {
    535    "cell_type": "markdown",
    536    "metadata": {},
    537    "source": [
    538     "For each model, the following plot shows the mean prediction and a band of two standard deviations around the mean for both models in the upper panel, and the bias-variance-noise breakdown based on the values for the true function in the bottom panel:"
    539    ]
    540   },
    541   {
    542    "cell_type": "code",
    543    "execution_count": 18,
    544    "metadata": {
    545     "ExecuteTime": {
    546      "end_time": "2018-11-09T16:08:56.085451Z",
    547      "start_time": "2018-11-09T16:08:50.487371Z"
    548     }
    549    },
    550    "outputs": [
    551     {
    552      "data": {
    553       "image/png": "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\n",
    554       "text/plain": [
    555        "<Figure size 1008x504 with 4 Axes>"
    556       ]
    557      },
    558      "metadata": {},
    559      "output_type": "display_data"
    560     }
    561    ],
    562    "source": [
    563     "fig, axes = plt.subplots(ncols=2, nrows=2, figsize=(14, 7), sharex=True, sharey='row')\n",
    564     "axes = axes.flatten()\n",
    565     "idx = pd.IndexSlice\n",
    566     "\n",
    567     "for i, (model, data) in enumerate(predictions.items()):\n",
    568     "    mean, std = data.mean(1), data.std(1).mul(2)\n",
    569     "    (pd.DataFrame([mean.sub(std), mean, mean.add(std)]).T\n",
    570     "     .sort_index()\n",
    571     "     .plot(style=['k--', 'k-', 'k--'], ax=axes[i], lw=1, legend=False))\n",
    572     "    (data.stack().reset_index()\n",
    573     "     .rename(columns={'level_0': 'x', 0: 'y'})\n",
    574     "     .plot.scatter(x='x', y='y', ax=axes[i], alpha=.02, s=2, color='r', title=model))\n",
    575     "    r = result[model]\n",
    576     "    m = r.mean()\n",
    577     "    kwargs = {'transform': axes[i].transAxes, 'fontsize':10}\n",
    578     "    axes[i].text(x=.8, y=.9, s=f'Bias: {m.bias:.2%}', **kwargs)\n",
    579     "    axes[i].text(x=.75, y=.8, s=f'Variance: {m.variance:.2%}', **kwargs)\n",
    580     "    \n",
    581     "    (r.drop('error', axis=1).sort_index()\n",
    582     "     .rename(columns=str.capitalize)\n",
    583     "     .plot(ax=axes[i+2], lw=1, legend=False, stacked=True))\n",
    584     "\n",
    585     "axes[-1].legend(fontsize=8)\n",
    586     "fig.suptitle('Bias-Variance Breakdown', fontsize=18)\n",
    587     "fig.tight_layout()\n",
    588     "fig.subplots_adjust(top=.9)\n",
    589     "fig.savefig('bias_variance_bagging', dpi=600);"
    590    ]
    591   },
    592   {
    593    "cell_type": "markdown",
    594    "metadata": {},
    595    "source": [
    596     "## Random Forests"
    597    ]
    598   },
    599   {
    600    "cell_type": "markdown",
    601    "metadata": {},
    602    "source": [
    603     "### Classifier"
    604    ]
    605   },
    606   {
    607    "cell_type": "markdown",
    608    "metadata": {},
    609    "source": [
    610     "### Cross-Validation with default settings"
    611    ]
    612   },
    613   {
    614    "cell_type": "code",
    615    "execution_count": 19,
    616    "metadata": {
    617     "ExecuteTime": {
    618      "end_time": "2018-11-09T16:03:40.040494Z",
    619      "start_time": "2018-11-09T16:03:40.036872Z"
    620     }
    621    },
    622    "outputs": [],
    623    "source": [
    624     "rf_clf = RandomForestClassifier(n_estimators=200,             # will change from 10 to 100 in version 0.22 \n",
    625     "                                criterion='gini', \n",
    626     "                                max_depth=None, \n",
    627     "                                min_samples_split=2, \n",
    628     "                                min_samples_leaf=1, \n",
    629     "                                min_weight_fraction_leaf=0.0, \n",
    630     "                                max_features='auto',\n",
    631     "                                max_leaf_nodes=None, \n",
    632     "                                min_impurity_decrease=0.0, \n",
    633     "                                min_impurity_split=None, \n",
    634     "                                bootstrap=True, \n",
    635     "                                oob_score=True, \n",
    636     "                                n_jobs=-1,\n",
    637     "                                random_state=42, \n",
    638     "                                verbose=1)"
    639    ]
    640   },
    641   {
    642    "cell_type": "code",
    643    "execution_count": 20,
    644    "metadata": {
    645     "ExecuteTime": {
    646      "end_time": "2018-11-09T16:03:40.860282Z",
    647      "start_time": "2018-11-09T16:03:40.853905Z"
    648     }
    649    },
    650    "outputs": [],
    651    "source": [
    652     "cv = OneStepTimeSeriesSplit(n_splits=12, test_period_length=1, shuffle=True)"
    653    ]
    654   },
    655   {
    656    "cell_type": "code",
    657    "execution_count": 21,
    658    "metadata": {
    659     "ExecuteTime": {
    660      "end_time": "2018-11-09T16:07:33.157682Z",
    661      "start_time": "2018-11-09T16:03:44.489874Z"
    662     }
    663    },
    664    "outputs": [
    665     {
    666      "name": "stderr",
    667      "output_type": "stream",
    668      "text": [
    669       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
    670       "[Parallel(n_jobs=-1)]: Done  10 out of  12 | elapsed:  4.4min remaining:   53.1s\n",
    671       "[Parallel(n_jobs=-1)]: Done  12 out of  12 | elapsed:  4.4min finished\n"
    672      ]
    673     }
    674    ],
    675    "source": [
    676     "cv_score = cross_val_score(estimator=rf_clf,\n",
    677     "                           X=X,\n",
    678     "                           y=y_binary,\n",
    679     "                           scoring='roc_auc',\n",
    680     "                           cv=cv,\n",
    681     "                           n_jobs=-1,\n",
    682     "                           verbose=1)"
    683    ]
    684   },
    685   {
    686    "cell_type": "code",
    687    "execution_count": 22,
    688    "metadata": {
    689     "ExecuteTime": {
    690      "end_time": "2018-11-09T16:07:33.163326Z",
    691      "start_time": "2018-11-09T16:07:33.159915Z"
    692     }
    693    },
    694    "outputs": [
    695     {
    696      "data": {
    697       "text/plain": [
    698        "0.521467856810403"
    699       ]
    700      },
    701      "execution_count": 22,
    702      "metadata": {},
    703      "output_type": "execute_result"
    704     }
    705    ],
    706    "source": [
    707     "np.mean(cv_score)"
    708    ]
    709   },
    710   {
    711    "cell_type": "markdown",
    712    "metadata": {},
    713    "source": [
    714     "### Parameter Tuning"
    715    ]
    716   },
    717   {
    718    "cell_type": "markdown",
    719    "metadata": {},
    720    "source": [
    721     "The key configuration parameters include the various hyperparameters for the individual decision trees introduced in the notebook [decision_trees](01_decision_trees.ipynb). \n",
    722     "\n",
    723     "The following tables lists additional options for the two `RandomForest` classes:"
    724    ]
    725   },
    726   {
    727    "cell_type": "markdown",
    728    "metadata": {},
    729    "source": [
    730     "| Keyword      | Default | Description                                                                                                                |\n",
    731     "|--------------|---------|----------------------------------------------------------------------------------------------------------------------------|\n",
    732     "| bootstrap    | True    | Bootstrap samples during training                                                                                          |\n",
    733     "| n_estimators | 10      | # trees in the forest.                                                                                                     |\n",
    734     "| oob_score    | False   | Use out-of-bag samples to estimate the R2 on unseen data                                                                   |\n",
    735     "| warm_start   | False   | Reuse result of previous call to continue training and add more trees to the ensemble, otherwise, train a whole new forest |"
    736    ]
    737   },
    738   {
    739    "cell_type": "markdown",
    740    "metadata": {},
    741    "source": [
    742     "- The `bootstrap` parameter activates in the preceding bagging algorithm outline, which in turn enables the computation of the out-of-bag score (oob_score) that estimates the generalization accuracy using samples not included in the bootstrap sample used to train a given tree (see next section for detail). \n",
    743     "- The `n_estimators` parameter defines the number of trees to be grown as part of the forest. Larger forests perform better, but also take more time to build. It is important to monitor the cross-validation error as a function of the number of base learners to identify when the marginal reduction of the prediction error declines and the cost of additional training begins to outweigh the benefits.\n",
    744     "- The `max_features` parameter controls the size of the randomly selected feature subsets available when learning a new decision rule and split a node. A lower value reduces the correlation of the trees and, thus, the ensemble's variance, but may also increase the bias. Good starting values are `n_features` (the number of training features) for regression problems and `sqrt(n_features)` for classification problems, but will depend on the relationships among features and should be optimized using cross-validation."
    745    ]
    746   },
    747   {
    748    "cell_type": "markdown",
    749    "metadata": {},
    750    "source": [
    751     "Random forests are designed to contain deep fully-grown trees, which can be created using `max_depth=None` and `min_samples_split=2`. However, these values are not necessarily optimal, especially for high-dimensional data with many samples and, consequently, potentially very deep trees that can become very computationally-, and memory-, intensive.\n",
    752     "\n",
    753     "The `RandomForest` class provided by sklearn support parallel training and prediction by setting the n_jobs parameter to the k number of jobs to run on different cores. The -1 value uses all available cores. The overhead of interprocess communication may limit the speedup from being linear so that k jobs may take more than 1/k the time of a single job. Nonetheless, the speedup is often quite significant for large forests or deep individual trees that may take a meaningful amount of time to train when the data is large, and split evaluation becomes costly.\n",
    754     "\n",
    755     "As always, the best parameter configuration should be identified using cross-validation. The following steps illustrate the process:"
    756    ]
    757   },
    758   {
    759    "cell_type": "markdown",
    760    "metadata": {},
    761    "source": [
    762     "#### Define Parameter Grid"
    763    ]
    764   },
    765   {
    766    "cell_type": "code",
    767    "execution_count": 23,
    768    "metadata": {
    769     "ExecuteTime": {
    770      "end_time": "2018-11-09T16:50:27.473681Z",
    771      "start_time": "2018-11-09T16:50:27.469346Z"
    772     }
    773    },
    774    "outputs": [],
    775    "source": [
    776     "param_grid = {'n_estimators': [200, 400],\n",
    777     "              'max_depth': [10, 15, 20],\n",
    778     "              'min_samples_leaf': [50, 100],\n",
    779     "              }"
    780    ]
    781   },
    782   {
    783    "cell_type": "markdown",
    784    "metadata": {},
    785    "source": [
    786     "#### Instantiate GridSearchCV"
    787    ]
    788   },
    789   {
    790    "cell_type": "markdown",
    791    "metadata": {},
    792    "source": [
    793     "We will use 10-fold custom cross-validation and populate the parameter grid with values for the key configuration settings:"
    794    ]
    795   },
    796   {
    797    "cell_type": "code",
    798    "execution_count": 24,
    799    "metadata": {
    800     "ExecuteTime": {
    801      "end_time": "2018-11-09T16:50:31.144810Z",
    802      "start_time": "2018-11-09T16:50:31.141653Z"
    803     }
    804    },
    805    "outputs": [],
    806    "source": [
    807     "gridsearch_clf = GridSearchCV(estimator=rf_clf,\n",
    808     "                              param_grid=param_grid,\n",
    809     "                              scoring='roc_auc',\n",
    810     "                              n_jobs=-1,\n",
    811     "                              cv=cv,\n",
    812     "                              refit=True,\n",
    813     "                              return_train_score=True,\n",
    814     "                              verbose=1)"
    815    ]
    816   },
    817   {
    818    "cell_type": "markdown",
    819    "metadata": {},
    820    "source": [
    821     "#### Fit"
    822    ]
    823   },
    824   {
    825    "cell_type": "code",
    826    "execution_count": 25,
    827    "metadata": {
    828     "ExecuteTime": {
    829      "end_time": "2018-11-09T17:35:06.706441Z",
    830      "start_time": "2018-11-09T16:50:35.372593Z"
    831     }
    832    },
    833    "outputs": [
    834     {
    835      "name": "stdout",
    836      "output_type": "stream",
    837      "text": [
    838       "Fitting 12 folds for each of 12 candidates, totalling 144 fits\n"
    839      ]
    840     },
    841     {
    842      "name": "stderr",
    843      "output_type": "stream",
    844      "text": [
    845       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
    846       "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:  8.8min\n",
    847       "[Parallel(n_jobs=-1)]: Done 144 out of 144 | elapsed: 46.3min finished\n",
    848       "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
    849       "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    3.2s\n",
    850       "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:   15.5s\n",
    851       "[Parallel(n_jobs=-1)]: Done 400 out of 400 | elapsed:   32.9s finished\n"
    852      ]
    853     },
    854     {
    855      "data": {
    856       "text/plain": [
    857        "GridSearchCV(cv=<__main__.OneStepTimeSeriesSplit object at 0x7f37b3539b70>,\n",
    858        "       error_score='raise-deprecating',\n",
    859        "       estimator=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
    860        "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
    861        "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
    862        "            min_samples_leaf=1, min_samples_split=2,\n",
    863        "            min_weight_fraction_leaf=0.0, n_estimators=200, n_jobs=-1,\n",
    864        "            oob_score=True, random_state=42, verbose=1, warm_start=False),\n",
    865        "       fit_params=None, iid='warn', n_jobs=-1,\n",
    866        "       param_grid={'n_estimators': [200, 400], 'max_depth': [10, 15, 20], 'min_samples_leaf': [50, 100]},\n",
    867        "       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
    868        "       scoring='roc_auc', verbose=1)"
    869       ]
    870      },
    871      "execution_count": 25,
    872      "metadata": {},
    873      "output_type": "execute_result"
    874     }
    875    ],
    876    "source": [
    877     "gridsearch_clf.fit(X=X, y=y_binary)"
    878    ]
    879   },
    880   {
    881    "cell_type": "markdown",
    882    "metadata": {},
    883    "source": [
    884     "#### Persist Result"
    885    ]
    886   },
    887   {
    888    "cell_type": "code",
    889    "execution_count": 26,
    890    "metadata": {
    891     "ExecuteTime": {
    892      "end_time": "2018-11-09T19:45:06.562351Z",
    893      "start_time": "2018-11-09T19:45:06.546117Z"
    894     }
    895    },
    896    "outputs": [
    897     {
    898      "data": {
    899       "text/plain": [
    900        "['gridsearch_clf.joblib']"
    901       ]
    902      },
    903      "execution_count": 26,
    904      "metadata": {},
    905      "output_type": "execute_result"
    906     }
    907    ],
    908    "source": [
    909     "joblib.dump(gridsearch_clf, 'gridsearch_clf.joblib') "
    910    ]
    911   },
    912   {
    913    "cell_type": "code",
    914    "execution_count": 27,
    915    "metadata": {
    916     "ExecuteTime": {
    917      "end_time": "2018-11-09T19:38:47.318454Z",
    918      "start_time": "2018-11-09T19:38:47.313371Z"
    919     }
    920    },
    921    "outputs": [
    922     {
    923      "data": {
    924       "text/plain": [
    925        "{'max_depth': 20, 'min_samples_leaf': 50, 'n_estimators': 400}"
    926       ]
    927      },
    928      "execution_count": 27,
    929      "metadata": {},
    930      "output_type": "execute_result"
    931     }
    932    ],
    933    "source": [
    934     "gridsearch_clf.best_params_"
    935    ]
    936   },
    937   {
    938    "cell_type": "code",
    939    "execution_count": 28,
    940    "metadata": {
    941     "ExecuteTime": {
    942      "end_time": "2018-11-09T19:38:47.575311Z",
    943      "start_time": "2018-11-09T19:38:47.565564Z"
    944     }
    945    },
    946    "outputs": [
    947     {
    948      "data": {
    949       "text/plain": [
    950        "0.5269293386898619"
    951       ]
    952      },
    953      "execution_count": 28,
    954      "metadata": {},
    955      "output_type": "execute_result"
    956     }
    957    ],
    958    "source": [
    959     "gridsearch_clf.best_score_"
    960    ]
    961   },
    962   {
    963    "cell_type": "markdown",
    964    "metadata": {},
    965    "source": [
    966     "### Feature Importance"
    967    ]
    968   },
    969   {
    970    "cell_type": "markdown",
    971    "metadata": {},
    972    "source": [
    973     "A random forest ensemble may contain hundreds of individual trees, but it is still possible to obtain an overall summary measure of feature importance from bagged models.\n",
    974     "\n",
    975     "For a given feature, the importance score is the total reduction in the objective function's value, which results from splits based on this feature, averaged over all trees. Since the objective function takes into account how many features are affected by a split, this measure is implicitly a weighted average so that features used near the top of a tree will get higher scores due to the larger number of observations contained in the much smaller number of available nodes. By averaging over many trees grown in a randomized fashion, the feature importance estimate loses some variance and becomes more accurate.\n",
    976     "\n",
    977     "The computation differs for classification and regression trees based on the different objectives used to learn the decision rules and is measured in terms of the mean square error for regression trees and the Gini index or entropy for classification trees.\n",
    978     "\n",
    979     "`sklearn` further normalizes the feature-importance measure so that it sums up to 1. Feature importance thus computed is also used for feature selection as an alternative to the mutual information measures we saw in Chapter 6, The Machine Learning Process (see SelectFromModel in the sklearn.feature_selection module).\n",
    980     "In our example, the importance values for the top-20 features are as shown here:"
    981    ]
    982   },
    983   {
    984    "cell_type": "code",
    985    "execution_count": 29,
    986    "metadata": {
    987     "ExecuteTime": {
    988      "end_time": "2018-11-09T19:47:04.790044Z",
    989      "start_time": "2018-11-09T19:47:04.512161Z"
    990     }
    991    },
    992    "outputs": [
    993     {
    994      "data": {
    995       "image/png": "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\n",
    996       "text/plain": [
    997        "<Figure size 432x288 with 1 Axes>"
    998       ]
    999      },
   1000      "metadata": {},
   1001      "output_type": "display_data"
   1002     }
   1003    ],
   1004    "source": [
   1005     "(pd.Series(gridsearch_clf.best_estimator_.feature_importances_, \n",
   1006     "           index=X.columns)\n",
   1007     " .sort_values(ascending=False)\n",
   1008     " .iloc[:20]\n",
   1009     " .sort_values()\n",
   1010     " .plot.barh());"
   1011    ]
   1012   },
   1013   {
   1014    "cell_type": "markdown",
   1015    "metadata": {},
   1016    "source": [
   1017     "### Regression"
   1018    ]
   1019   },
   1020   {
   1021    "cell_type": "code",
   1022    "execution_count": 37,
   1023    "metadata": {
   1024     "ExecuteTime": {
   1025      "end_time": "2018-11-09T17:35:06.745906Z",
   1026      "start_time": "2018-11-09T17:14:25.425Z"
   1027     }
   1028    },
   1029    "outputs": [],
   1030    "source": [
   1031     "rf_reg = RandomForestRegressor(n_estimators=10, \n",
   1032     "                                max_depth=None, \n",
   1033     "                                min_samples_split=2, \n",
   1034     "                                min_samples_leaf=1, \n",
   1035     "                                min_weight_fraction_leaf=0.0, \n",
   1036     "                                max_features='auto', \n",
   1037     "                                max_leaf_nodes=None, \n",
   1038     "                                min_impurity_decrease=0.0, \n",
   1039     "                                min_impurity_split=None, \n",
   1040     "                                bootstrap=True, \n",
   1041     "                                oob_score=False, \n",
   1042     "                                n_jobs=-1, \n",
   1043     "                                random_state=None, \n",
   1044     "                                verbose=0, \n",
   1045     "                                warm_start=False)"
   1046    ]
   1047   },
   1048   {
   1049    "cell_type": "code",
   1050    "execution_count": 38,
   1051    "metadata": {
   1052     "ExecuteTime": {
   1053      "end_time": "2018-11-09T17:35:06.746456Z",
   1054      "start_time": "2018-11-09T17:14:28.567Z"
   1055     }
   1056    },
   1057    "outputs": [],
   1058    "source": [
   1059     "gridsearch_reg = GridSearchCV(estimator=rf_reg,\n",
   1060     "                              param_grid=param_grid,\n",
   1061     "                              scoring='neg_mean_squared_error',\n",
   1062     "                              n_jobs=-1,\n",
   1063     "                              cv=cv,\n",
   1064     "                              refit=True,\n",
   1065     "                              return_train_score=True,\n",
   1066     "                              verbose=1)"
   1067    ]
   1068   },
   1069   {
   1070    "cell_type": "code",
   1071    "execution_count": 39,
   1072    "metadata": {},
   1073    "outputs": [
   1074     {
   1075      "data": {
   1076       "text/plain": [
   1077        "(171162, 60)"
   1078       ]
   1079      },
   1080      "execution_count": 39,
   1081      "metadata": {},
   1082      "output_type": "execute_result"
   1083     }
   1084    ],
   1085    "source": [
   1086     "X.shape"
   1087    ]
   1088   },
   1089   {
   1090    "cell_type": "code",
   1091    "execution_count": 40,
   1092    "metadata": {},
   1093    "outputs": [
   1094     {
   1095      "data": {
   1096       "text/plain": [
   1097        "(171162,)"
   1098       ]
   1099      },
   1100      "execution_count": 40,
   1101      "metadata": {},
   1102      "output_type": "execute_result"
   1103     }
   1104    ],
   1105    "source": [
   1106     "y.shape"
   1107    ]
   1108   },
   1109   {
   1110    "cell_type": "code",
   1111    "execution_count": 41,
   1112    "metadata": {
   1113     "ExecuteTime": {
   1114      "end_time": "2018-11-09T17:35:06.746990Z",
   1115      "start_time": "2018-11-09T17:14:33.748Z"
   1116     }
   1117    },
   1118    "outputs": [
   1119     {
   1120      "name": "stdout",
   1121      "output_type": "stream",
   1122      "text": [
   1123       "Fitting 12 folds for each of 12 candidates, totalling 144 fits\n"
   1124      ]
   1125     },
   1126     {
   1127      "name": "stderr",
   1128      "output_type": "stream",
   1129      "text": [
   1130       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
   1131       "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed: 44.6min\n",
   1132       "[Parallel(n_jobs=-1)]: Done 144 out of 144 | elapsed: 245.6min finished\n"
   1133      ]
   1134     },
   1135     {
   1136      "data": {
   1137       "text/plain": [
   1138        "GridSearchCV(cv=<__main__.OneStepTimeSeriesSplit object at 0x7f37b3539b70>,\n",
   1139        "       error_score='raise-deprecating',\n",
   1140        "       estimator=RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
   1141        "           max_features='auto', max_leaf_nodes=None,\n",
   1142        "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
   1143        "           min_samples_leaf=1, min_samples_split=2,\n",
   1144        "           min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=-1,\n",
   1145        "           oob_score=False, random_state=None, verbose=0, warm_start=False),\n",
   1146        "       fit_params=None, iid='warn', n_jobs=-1,\n",
   1147        "       param_grid={'n_estimators': [200, 400], 'max_depth': [10, 15, 20], 'min_samples_leaf': [50, 100]},\n",
   1148        "       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
   1149        "       scoring='neg_mean_squared_error', verbose=1)"
   1150       ]
   1151      },
   1152      "execution_count": 41,
   1153      "metadata": {},
   1154      "output_type": "execute_result"
   1155     }
   1156    ],
   1157    "source": [
   1158     "gridsearch_reg.fit(X=X, y=y)"
   1159    ]
   1160   },
   1161   {
   1162    "cell_type": "code",
   1163    "execution_count": 42,
   1164    "metadata": {
   1165     "ExecuteTime": {
   1166      "end_time": "2018-11-09T17:35:06.749649Z",
   1167      "start_time": "2018-11-09T17:14:49.048Z"
   1168     }
   1169    },
   1170    "outputs": [
   1171     {
   1172      "data": {
   1173       "text/plain": [
   1174        "['rf_reg_gridsearch.joblib']"
   1175       ]
   1176      },
   1177      "execution_count": 42,
   1178      "metadata": {},
   1179      "output_type": "execute_result"
   1180     }
   1181    ],
   1182    "source": [
   1183     "joblib.dump(gridsearch_reg, 'rf_reg_gridsearch.joblib') "
   1184    ]
   1185   },
   1186   {
   1187    "cell_type": "code",
   1188    "execution_count": 43,
   1189    "metadata": {
   1190     "ExecuteTime": {
   1191      "end_time": "2018-11-09T17:35:06.750211Z",
   1192      "start_time": "2018-11-09T17:14:49.523Z"
   1193     }
   1194    },
   1195    "outputs": [
   1196     {
   1197      "data": {
   1198       "text/plain": [
   1199        "{'max_depth': 10, 'min_samples_leaf': 100, 'n_estimators': 200}"
   1200       ]
   1201      },
   1202      "execution_count": 43,
   1203      "metadata": {},
   1204      "output_type": "execute_result"
   1205     }
   1206    ],
   1207    "source": [
   1208     "gridsearch_reg.best_params_"
   1209    ]
   1210   },
   1211   {
   1212    "cell_type": "code",
   1213    "execution_count": 44,
   1214    "metadata": {
   1215     "ExecuteTime": {
   1216      "end_time": "2018-11-01T01:45:35.099508Z",
   1217      "start_time": "2018-10-31T22:34:01.660Z"
   1218     }
   1219    },
   1220    "outputs": [
   1221     {
   1222      "data": {
   1223       "text/plain": [
   1224        "'7.16%'"
   1225       ]
   1226      },
   1227      "execution_count": 44,
   1228      "metadata": {},
   1229      "output_type": "execute_result"
   1230     }
   1231    ],
   1232    "source": [
   1233     "f'{np.sqrt(-gridsearch_reg.best_score_):.2%}'"
   1234    ]
   1235   },
   1236   {
   1237    "cell_type": "code",
   1238    "execution_count": 45,
   1239    "metadata": {
   1240     "ExecuteTime": {
   1241      "end_time": "2018-11-01T01:45:35.100421Z",
   1242      "start_time": "2018-10-31T22:34:01.671Z"
   1243     }
   1244    },
   1245    "outputs": [
   1246     {
   1247      "data": {
   1248       "text/plain": [
   1249        "'6.55%'"
   1250       ]
   1251      },
   1252      "execution_count": 45,
   1253      "metadata": {},
   1254      "output_type": "execute_result"
   1255     }
   1256    ],
   1257    "source": [
   1258     "f'{regression_benchmark():.2%}'"
   1259    ]
   1260   }
   1261  ],
   1262  "metadata": {
   1263   "kernelspec": {
   1264    "display_name": "Python 3",
   1265    "language": "python",
   1266    "name": "python3"
   1267   },
   1268   "language_info": {
   1269    "codemirror_mode": {
   1270     "name": "ipython",
   1271     "version": 3
   1272    },
   1273    "file_extension": ".py",
   1274    "mimetype": "text/x-python",
   1275    "name": "python",
   1276    "nbconvert_exporter": "python",
   1277    "pygments_lexer": "ipython3",
   1278    "version": "3.7.0"
   1279   },
   1280   "toc": {
   1281    "base_numbering": 1,
   1282    "nav_menu": {},
   1283    "number_sections": true,
   1284    "sideBar": true,
   1285    "skip_h1_title": false,
   1286    "title_cell": "Table of Contents",
   1287    "title_sidebar": "Contents",
   1288    "toc_cell": false,
   1289    "toc_position": {
   1290     "height": "calc(100% - 180px)",
   1291     "left": "10px",
   1292     "top": "150px",
   1293     "width": "281.225px"
   1294    },
   1295    "toc_section_display": true,
   1296    "toc_window_display": true
   1297   }
   1298  },
   1299  "nbformat": 4,
   1300  "nbformat_minor": 2
   1301 }