Raw File
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CNTK 106:  Part B - Time series prediction with LSTM (IOT Data)\n",
    "\n",
    "In [part A of this tutorial](CNTK_106A_LSTM_Timeseries_with_Simulated_Data.ipynb) we developed a simple LSTM network to predict future values in a time series. In part B we want to use the model on some real world internet-of-things ([IOT](https://en.wikipedia.org/wiki/Internet_of_things)) data. As an example we want to predict the daily output of a solar panel base on the initial readings of the day. \n",
    "\n",
    "[Solar power forecasting](https://en.wikipedia.org/wiki/Solar_power_forecasting) is a challenging and important problem. The solar energy generation forecasting problem is closely linked to the problem of weather variables forecasting. Indeed, this problem is usually split into two parts, on one hand focusing on the forecasting of solar PV or any other meteorological variable and on the other hand estimating the amount of energy that a concrete power plant will produce with the estimated meteorological resource. In general, the way to deal with this difficult problem is usually related to the spatial and temporal scales we are interested in. This tutorial focusses on a simplified forecasting model using previously generated data from solar panel to predict the future. \n",
    "\n",
    "**Goal**\n",
    "\n",
    "Using historic daily production of a solar panel, we want to predict the total power production of the solar panel array for a day. We will be using the LSTM based time series prediction model developed in part A to predict the daily output of a solar panel based on the initial readings of the day. \n",
    "\n",
    "![](https://www.cntk.ai/jup/rooftop-solar-power.jpg)\n",
    "\n",
    "We train the model with historical data of the solar panel. In our example we want to predict the total power production of the solar panel array for the day starting with the initial readings of the day. We start predicting after the first 2 readings and adjust the prediction with each new reading.\n",
    "\n",
    "In this tutorial, we will use the LSTM model introduced in the CNTK 106A. This tutorial has the following sub-sections:\n",
    "- Setup\n",
    "- Data generation\n",
    "- LSTM network modeling\n",
    "- Model training and evaluation\n",
    "\n",
    "For more details on how LSTMs work, see [this excellent post](http://colah.github.io/posts/2015-08-Understanding-LSTMs)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "We need a few imports and constants throughout the tutorial that we define here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "import math\n",
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import random\n",
    "import time\n",
    "\n",
    "import cntk as C\n",
    "\n",
    "try:\n",
    "    from urllib.request import urlretrieve\n",
    "except ImportError:\n",
    "    from urllib import urlretrieve\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "import cntk.tests.test_utils\n",
    "cntk.tests.test_utils.set_device_from_pytest_env() # (only needed for our build system)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# to make things reproduceable, seed random\n",
    "np.random.seed(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are two run modes:\n",
    "- *Fast mode*: `isFast` is set to `True`. This is the default mode for the notebooks, which means we train for fewer iterations or train / test on limited data. This ensures functional correctness of the notebook though the models produced are far from what a completed training would produce.\n",
    "\n",
    "- *Slow mode*: We recommend the user to set this flag to `False` once the user has gained familiarity with the notebook content and wants to gain insight from running the notebooks for a longer period with different parameters for training. \n",
    "\n",
    "For *Fast mode* we train the model for 100 epochs and results have low accuracy but is good enough for development. The model yields good accuracy after 1000-2000 epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "isFast = True\n",
    "\n",
    "# we need around 2000 epochs to see good accuracy. For testing 100 epochs will do.\n",
    "EPOCHS = 100 if isFast else 2000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data generation\n",
    "\n",
    "Our solar panel, emits two measures at every 30 min interval:\n",
    "- `solar.current` is the current production in Watt\n",
    "- `solar.total` is the total produced for the day so far in Watt/hour\n",
    "\n",
    "Our prediction approach involves starting with the first 2 initial readings of the day. Based on these readings we start predicting and adjust the prediction with each new reading. The training data we are going to use comes as a csv file and has the following format:\n",
    "\n",
    ">```\n",
    "time,solar.current,solar.total\n",
    "7am,6.3,1.7\n",
    "7:30am,44.3,11.4\n",
    "...\n",
    ">```\n",
    "\n",
    "The training dataset we use contains data captured for 3 years and can be found [here](https://guschmueds.blob.core.windows.net/datasets/solar.csv). \n",
    "The dataset is not pre-processed: it is raw data and contains smaller gaps and errors (like a panel failed to report)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pre-processing\n",
    "Most of the code in this example is related to data preparation. Thankfully the pandas library make this easy.\n",
    "\n",
    "`generate_solar_data()` function performs the following tasks:\n",
    "- read raw data into a pandas dataframe, \n",
    "- normalize the data, \n",
    "- groups by day and \n",
    "- append the columns \"solar.current.max\" and \"solar.total.max\", and\n",
    "- generates the sequences for each day.\n",
    "\n",
    "*Sequence generation*: All sequences are concatenated into a single list of sequences. There is *no more notion of timestamp* in our train input and **only** the sequences matter.\n",
    "\n",
    "**Note** if we have less than 8 datapoints for a day we skip over the day assuming something is missing in the raw data. If we get more than 14 data points in a day we truncate the readings.\n",
    "\n",
    "### Training / Testing / Validation data preparation\n",
    "We start by reading the csv file for use with CNTK.  The raw data is sorted by time and we should randomize it before splitting into training, validation and test datasets but this would make it impractical to visualize results in the tutorial. Hence, we split the dataset in the following manner: pick in sequence, 8 values for training, 1 for validation and 1 for test until there is no more data. This will spread training, validation and test datasets across the full timeline while preserving time order.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def generate_solar_data(input_url, time_steps, normalize=1, val_size=0.1, test_size=0.1):\n",
    "    \"\"\"\n",
    "    generate sequences to feed to rnn based on data frame with solar panel data\n",
    "    the csv has the format: time ,solar.current, solar.total\n",
    "     (solar.current is the current output in Watt, solar.total is the total production\n",
    "      for the day so far in Watt hours)\n",
    "    \"\"\"\n",
    "    # try to find the data file local. If it doesn't exists download it.\n",
    "    cache_path = os.path.join(\"data\", \"iot\")\n",
    "    cache_file = os.path.join(cache_path, \"solar.csv\")\n",
    "    if not os.path.exists(cache_path):\n",
    "        os.makedirs(cache_path)\n",
    "    if not os.path.exists(cache_file):\n",
    "        urlretrieve(input_url, cache_file)\n",
    "        print(\"downloaded data successfully from \", input_url)\n",
    "    else:\n",
    "        print(\"using cache for \", input_url)\n",
    "    \n",
    "    df = pd.read_csv(cache_file, index_col=\"time\", parse_dates=['time'], dtype=np.float32)\n",
    "    \n",
    "    df[\"date\"] = df.index.date\n",
    "    \n",
    "    # normalize data\n",
    "    df['solar.current'] /= normalize\n",
    "    df['solar.total'] /= normalize\n",
    "    \n",
    "    # group by day, find the max for a day and add a new column .max\n",
    "    grouped = df.groupby(df.index.date).max()\n",
    "    grouped.columns = [\"solar.current.max\", \"solar.total.max\", \"date\"]\n",
    "\n",
    "    # merge continuous readings and daily max values into a single frame\n",
    "    df_merged = pd.merge(df, grouped, right_index=True, on=\"date\")\n",
    "    df_merged = df_merged[[\"solar.current\", \"solar.total\",\n",
    "                           \"solar.current.max\", \"solar.total.max\"]]\n",
    "    # we group by day so we can process a day at a time.\n",
    "    grouped = df_merged.groupby(df_merged.index.date)\n",
    "    per_day = []\n",
    "    for _, group in grouped:\n",
    "        per_day.append(group)\n",
    "\n",
    "    # split the dataset into train, validatation and test sets on day boundaries\n",
    "    val_size = int(len(per_day) * val_size)\n",
    "    test_size = int(len(per_day) * test_size)\n",
    "    next_val = 0\n",
    "    next_test = 0\n",
    "\n",
    "    result_x = {\"train\": [], \"val\": [], \"test\": []}\n",
    "    result_y = {\"train\": [], \"val\": [], \"test\": []}    \n",
    "\n",
    "    # generate sequences a day at a time\n",
    "    for i, day in enumerate(per_day):\n",
    "        # if we have less than 8 datapoints for a day we skip over the\n",
    "        # day assuming something is missing in the raw data\n",
    "        total = day[\"solar.total\"].values\n",
    "        if len(total) < 8:\n",
    "            continue\n",
    "        if i >= next_val:\n",
    "            current_set = \"val\"\n",
    "            next_val = i + int(len(per_day) / val_size)\n",
    "        elif i >= next_test:\n",
    "            current_set = \"test\"\n",
    "            next_test = i + int(len(per_day) / test_size)\n",
    "        else:\n",
    "            current_set = \"train\"\n",
    "        max_total_for_day = np.array(day[\"solar.total.max\"].values[0])\n",
    "        for j in range(2, len(total)):\n",
    "            result_x[current_set].append(total[0:j])\n",
    "            result_y[current_set].append([max_total_for_day])\n",
    "            if j >= time_steps:\n",
    "                break\n",
    "    # make result_y a numpy array\n",
    "    for ds in [\"train\", \"val\", \"test\"]:\n",
    "        result_y[ds] = np.array(result_y[ds])\n",
    "    return result_x, result_y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data caching\n",
    "For routine testing we would like to cache the data locally when available. If it is not available from the cache locations we shall download."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "using cache for  https://www.cntk.ai/jup/dat/solar.csv\n"
     ]
    }
   ],
   "source": [
    "# We keep upto 14 inputs from a day\n",
    "TIMESTEPS = 14\n",
    "\n",
    "# 20000 is the maximum total output in our dataset. We normalize all values with \n",
    "# this so our inputs are between 0.0 and 1.0 range.\n",
    "NORMALIZE = 20000\n",
    "\n",
    "X, Y = generate_solar_data(\"https://www.cntk.ai/jup/dat/solar.csv\", \n",
    "                           TIMESTEPS, normalize=NORMALIZE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Utility for data fetching**\n",
    "\n",
    "`next_batch()` yields the next batch for training. We use variable size sequences supported by CNTK and batches are a list of numpy arrays where the numpy arrays have variable length. \n",
    "\n",
    "A standard practice is to shuffle batches with each epoch. We don't do this here because we want to be able to graph the data that is easily interpretable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# process batches of 10 days\n",
    "BATCH_SIZE = TIMESTEPS * 10\n",
    "\n",
    "def next_batch(x, y, ds):\n",
    "    \"\"\"get the next batch for training\"\"\"\n",
    "\n",
    "    def as_batch(data, start, count):\n",
    "        return data[start:start + count]\n",
    "\n",
    "    for i in range(0, len(x[ds]), BATCH_SIZE):\n",
    "        yield as_batch(X[ds], i, BATCH_SIZE), as_batch(Y[ds], i, BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Understand the data format**\n",
    "\n",
    "You can now see the sequence we are going to feed to the LSTM. Note if we have less than 8 datapoints for a day we skip over the day assuming something is missing in the raw data. If we get more than 14 data points in a day we truncate the readings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([ 0.       ,  0.0006985], dtype=float32),\n",
       " array([ 0.       ,  0.0006985,  0.0033175], dtype=float32),\n",
       " array([ 0.       ,  0.0006985,  0.0033175,  0.010375 ], dtype=float32)]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X['train'][0:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.23899999],\n",
       "       [ 0.23899999],\n",
       "       [ 0.23899999]], dtype=float32)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y['train'][0:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Creation (LSTM network)\n",
    "\n",
    "Corresponding  to the maximum possible 14 data points in the input sequence, we model our network with 14 LSTM cells, 1 cell for each data point we take during the day. Since the input sequences can be between 8 and 14 data points per sequence, we take the advantage of CNTK support for variable sequences as input to a LSTM so we can feed our sequences as-is with no additional need for padding.\n",
    "\n",
    "The output of the neural network is the total output for the day and each sequence for a given day has the same total output.\n",
    "\n",
    "For example:\n",
    "```\n",
    "1.7,11.4 -> 10300\n",
    "1.7,11.4,67.5 -> 10300\n",
    "1.7,11.4,67.5,250.5 ... -> 10300\n",
    "1.7,11.4,67.5,250.5,573.5 -> 10300\n",
    "```\n",
    "\n",
    "The outputs from the LSTMs are fed into a dense layer and we randomly dropout 20% of the values to not overfit the model to the training set. The output of the dense layer becomes the prediction our model generates.\n",
    "\n",
    "Notice: We lost the timestamp altogether; in our model only the sequences of readings matter. \n",
    "\n",
    "Our LSTM model has the following design:\n",
    "![lstm](https://guschmueds.blob.core.windows.net/datasets/2.png)\n",
    "\n",
    "The network model is an exact translation of the network diagram above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Specify the internal-state dimensions of the LSTM cell\n",
    "H_DIMS = 15\n",
    "def create_model(x):\n",
    "    \"\"\"Create the model for time series prediction\"\"\"\n",
    "    with C.layers.default_options(initial_state = 0.1):\n",
    "        m = C.layers.Recurrence(C.layers.LSTM(H_DIMS))(x)\n",
    "        m = C.sequence.last(m)\n",
    "        m = C.layers.Dropout(0.2)(m)\n",
    "        m = C.layers.Dense(1)(m)\n",
    "        return m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training\n",
    "Before we can start training we need to bind our input variables for the model and define what optimizer we want to use. For this example we choose the `adam` optimizer. We choose `squared_error` as our loss function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# input sequences\n",
    "x = C.sequence.input_variable(1)\n",
    "\n",
    "# create the model\n",
    "z = create_model(x)\n",
    "\n",
    "# expected output (label), also the dynamic axes of the model output\n",
    "# is specified as the model of the label input\n",
    "l = C.input_variable(1, dynamic_axes=z.dynamic_axes, name=\"y\")\n",
    "\n",
    "# the learning rate\n",
    "learning_rate = 0.005\n",
    "lr_schedule = C.learning_rate_schedule(learning_rate, C.UnitType.minibatch)\n",
    "\n",
    "# loss function\n",
    "loss = C.squared_error(z, l)\n",
    "\n",
    "# use squared error to determine error for now\n",
    "error = C.squared_error(z, l)\n",
    "\n",
    "# use adam optimizer\n",
    "momentum_time_constant = C.momentum_as_time_constant_schedule(BATCH_SIZE / -math.log(0.9)) \n",
    "learner = C.fsadagrad(z.parameters, \n",
    "                      lr = lr_schedule, \n",
    "                      momentum = momentum_time_constant)\n",
    "trainer = C.Trainer(z, (loss, error), [learner])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Time to start training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 0, loss: 0.0768\n",
      "epoch: 10, loss: 0.0265\n",
      "epoch: 20, loss: 0.0158\n",
      "epoch: 30, loss: 0.0104\n",
      "epoch: 40, loss: 0.0082\n",
      "epoch: 50, loss: 0.0081\n",
      "epoch: 60, loss: 0.0088\n",
      "epoch: 70, loss: 0.0080\n",
      "epoch: 80, loss: 0.0082\n",
      "epoch: 90, loss: 0.0077\n",
      "Training took 171.8 sec\n"
     ]
    }
   ],
   "source": [
    "# training\n",
    "loss_summary = []\n",
    "\n",
    "start = time.time()\n",
    "for epoch in range(0, EPOCHS):\n",
    "    for x_batch, l_batch in next_batch(X, Y, \"train\"):\n",
    "        trainer.train_minibatch({x: x_batch, l: l_batch})\n",
    "        \n",
    "    if epoch % (EPOCHS / 10) == 0:\n",
    "        training_loss = trainer.previous_minibatch_loss_average\n",
    "        loss_summary.append(training_loss)\n",
    "        print(\"epoch: {}, loss: {:.4f}\".format(epoch, training_loss))\n",
    "\n",
    "print(\"Training took {:.1f} sec\".format(time.time() - start))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A look how the loss function shows how the model is converging:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEACAYAAAC9Gb03AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noTczyzkXejOznPv/uATLs5UwDXUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x13875d2b198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(loss_summary, label='training loss');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us validate the training validation and test dataset. We use mean squared error as measure which might be a little simplistic. A method that would define a ratio how many predictions have been inside a given tolerance would make a better measure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# validate\n",
    "def get_mse(X,Y,labeltxt):\n",
    "    result = 0.0\n",
    "    for x1, y1 in next_batch(X, Y, labeltxt):\n",
    "        eval_error = trainer.test_minibatch({x : x1, l : y1})\n",
    "        result += eval_error\n",
    "    return result/len(X[labeltxt])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mse for train: 0.000079\n",
      "mse for val: 0.000077\n"
     ]
    }
   ],
   "source": [
    "# Print the train and validation errors\n",
    "for labeltxt in [\"train\", \"val\"]:\n",
    "    print(\"mse for {}: {:.6f}\".format(labeltxt, get_mse(X, Y, labeltxt)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mse for test: 0.000076\n"
     ]
    }
   ],
   "source": [
    "# Print the test error\n",
    "labeltxt = \"test\"\n",
    "print(\"mse for {}: {:.6f}\".format(labeltxt, get_mse(X, Y, labeltxt)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize results\n",
    "\n",
    "Our model has been trained well, given the train, validation and test errors are in the same ball park. Predicted time series data renders well with visualization of the results. Let us take our newly created model, make predictions and plot them against the actual readings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtwAAAHfCAYAAABj1Q9aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x138768de978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict\n",
    "f, a = plt.subplots(2, 1, figsize=(12, 8))\n",
    "for j, ds in enumerate([\"val\", \"test\"]):\n",
    "    results = []\n",
    "    for x_batch, _ in next_batch(X, Y, ds):\n",
    "        pred = z.eval({x: x_batch})\n",
    "        results.extend(pred[:, 0])\n",
    "    # because we normalized the input data we need to multiply the prediction\n",
    "    # with SCALER to get the real values.\n",
    "    a[j].plot((Y[ds] * NORMALIZE).flatten(), label=ds + ' raw');\n",
    "    a[j].plot(np.array(results) * NORMALIZE, label=ds + ' pred');\n",
    "    a[j].legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we let the model train for 2000 epochs the predictions are close to the actual data and follow the right pattern."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "**Suggested activity**\n",
    "\n",
    "So what we do with this model? A practical application would be to generate alerts if the actual output is not in line with the prediction, for example if one of the panels is failing. The solar array that goes with our dataset has 16 panels. If we'd want to detect failure without generating false alerts, the accuracy of our prediction would need to be at least 1 - 1/16, around 94%. Our model is close to this but would most likely generate occasional false alerts.\n",
    "\n",
    "- Improve the model by training for more epochs.\n",
    "- Further preprocess the training set to smooth-out missing values\n",
    "- Try out more complex networks.\n",
    "\n",
    "However, in our experience with time series data one can achieve significant accuracy improvement comes from higher resolution training data, for example reading a data point every 5 minutes instead of every 30 minutes.\n",
    "\n",
    "We hope this tutorial gets you started on time series prediction with neural networks."
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}
back to top