synthetic_features_and_outliers.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "synthetic_features_and_outliers.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [
        "JndnmDMp66FL",
        "i5Ul3zf5QYvW",
        "jByCP8hDRZmM",
        "WvgxW0bUSC-c"
      ]
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "[View in Colaboratory](https://colab.research.google.com/github/chaiban/tutorials/blob/master/synthetic_features_and_outliers.ipynb)"
      ]
    },
    {
      "metadata": {
        "id": "JndnmDMp66FL",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "#### Copyright 2017 Google LLC."
      ]
    },
    {
      "metadata": {
        "id": "hMqWDc_m6rUC",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "4f3CKqFUqL2-",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Synthetic Features and Outliers"
      ]
    },
    {
      "metadata": {
        "id": "jnKgkN5fHbGy",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Learning Objectives:**\n",
        "  * Create a synthetic feature that is the ratio of two other features\n",
        "  * Use this new feature as an input to a linear regression model\n",
        "  * Improve the effectiveness of the model by identifying and clipping (removing) outliers out of the input data"
      ]
    },
    {
      "metadata": {
        "id": "VOpLo5dcHbG0",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Let's revisit our model from the previous First Steps with TensorFlow exercise. \n",
        "\n",
        "First, we'll import the California housing data into a *pandas* `DataFrame`:"
      ]
    },
    {
      "metadata": {
        "id": "S8gm6BpqRRuh",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Setup"
      ]
    },
    {
      "metadata": {
        "id": "9D8GgUovHbG0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "outputId": "a37fe8b1-2929-4fc4-b86b-3ec226326cd0"
      },
      "cell_type": "code",
      "source": [
        "import math\n",
        "\n",
        "from IPython import display\n",
        "from matplotlib import cm\n",
        "from matplotlib import gridspec\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import sklearn.metrics as metrics\n",
        "import tensorflow as tf\n",
        "from tensorflow.python.data import Dataset\n",
        "\n",
        "tf.logging.set_verbosity(tf.logging.ERROR)\n",
        "pd.options.display.max_rows = 10\n",
        "pd.options.display.float_format = '{:.1f}'.format\n",
        "\n",
        "california_housing_dataframe = pd.read_csv(\"https://storage.googleapis.com/mledu-datasets/california_housing_train.csv\", sep=\",\")\n",
        "\n",
        "california_housing_dataframe = california_housing_dataframe.reindex(\n",
        "    np.random.permutation(california_housing_dataframe.index))\n",
        "california_housing_dataframe[\"median_house_value\"] /= 1000.0\n",
        "california_housing_dataframe"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>longitude</th>\n",
              "      <th>latitude</th>\n",
              "      <th>housing_median_age</th>\n",
              "      <th>total_rooms</th>\n",
              "      <th>total_bedrooms</th>\n",
              "      <th>population</th>\n",
              "      <th>households</th>\n",
              "      <th>median_income</th>\n",
              "      <th>median_house_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>11923</th>\n",
              "      <td>-121.4</td>\n",
              "      <td>38.7</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1093.0</td>\n",
              "      <td>192.0</td>\n",
              "      <td>590.0</td>\n",
              "      <td>190.0</td>\n",
              "      <td>2.7</td>\n",
              "      <td>80.2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9232</th>\n",
              "      <td>-119.1</td>\n",
              "      <td>35.3</td>\n",
              "      <td>16.0</td>\n",
              "      <td>4201.0</td>\n",
              "      <td>786.0</td>\n",
              "      <td>1667.0</td>\n",
              "      <td>724.0</td>\n",
              "      <td>4.9</td>\n",
              "      <td>134.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15285</th>\n",
              "      <td>-122.3</td>\n",
              "      <td>37.9</td>\n",
              "      <td>49.0</td>\n",
              "      <td>2932.0</td>\n",
              "      <td>668.0</td>\n",
              "      <td>1361.0</td>\n",
              "      <td>608.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>147.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7025</th>\n",
              "      <td>-118.3</td>\n",
              "      <td>33.8</td>\n",
              "      <td>31.0</td>\n",
              "      <td>4464.0</td>\n",
              "      <td>991.0</td>\n",
              "      <td>2420.0</td>\n",
              "      <td>947.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>277.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9612</th>\n",
              "      <td>-119.4</td>\n",
              "      <td>36.4</td>\n",
              "      <td>20.0</td>\n",
              "      <td>1469.0</td>\n",
              "      <td>303.0</td>\n",
              "      <td>1031.0</td>\n",
              "      <td>259.0</td>\n",
              "      <td>1.7</td>\n",
              "      <td>48.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10554</th>\n",
              "      <td>-120.4</td>\n",
              "      <td>34.6</td>\n",
              "      <td>30.0</td>\n",
              "      <td>2265.0</td>\n",
              "      <td>512.0</td>\n",
              "      <td>1402.0</td>\n",
              "      <td>471.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>134.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13842</th>\n",
              "      <td>-122.0</td>\n",
              "      <td>37.6</td>\n",
              "      <td>6.0</td>\n",
              "      <td>2445.0</td>\n",
              "      <td>590.0</td>\n",
              "      <td>1189.0</td>\n",
              "      <td>573.0</td>\n",
              "      <td>3.9</td>\n",
              "      <td>301.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4555</th>\n",
              "      <td>-118.0</td>\n",
              "      <td>33.9</td>\n",
              "      <td>16.0</td>\n",
              "      <td>2676.0</td>\n",
              "      <td>391.0</td>\n",
              "      <td>1377.0</td>\n",
              "      <td>395.0</td>\n",
              "      <td>6.6</td>\n",
              "      <td>350.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1962</th>\n",
              "      <td>-117.3</td>\n",
              "      <td>33.1</td>\n",
              "      <td>14.0</td>\n",
              "      <td>2670.0</td>\n",
              "      <td>426.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>407.0</td>\n",
              "      <td>6.4</td>\n",
              "      <td>295.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14377</th>\n",
              "      <td>-122.1</td>\n",
              "      <td>37.4</td>\n",
              "      <td>16.0</td>\n",
              "      <td>1994.0</td>\n",
              "      <td>489.0</td>\n",
              "      <td>1173.0</td>\n",
              "      <td>472.0</td>\n",
              "      <td>4.2</td>\n",
              "      <td>266.4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>17000 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
              "11923     -121.4      38.7                35.0       1093.0           192.0   \n",
              "9232      -119.1      35.3                16.0       4201.0           786.0   \n",
              "15285     -122.3      37.9                49.0       2932.0           668.0   \n",
              "7025      -118.3      33.8                31.0       4464.0           991.0   \n",
              "9612      -119.4      36.4                20.0       1469.0           303.0   \n",
              "...          ...       ...                 ...          ...             ...   \n",
              "10554     -120.4      34.6                30.0       2265.0           512.0   \n",
              "13842     -122.0      37.6                 6.0       2445.0           590.0   \n",
              "4555      -118.0      33.9                16.0       2676.0           391.0   \n",
              "1962      -117.3      33.1                14.0       2670.0           426.0   \n",
              "14377     -122.1      37.4                16.0       1994.0           489.0   \n",
              "\n",
              "       population  households  median_income  median_house_value  \n",
              "11923       590.0       190.0            2.7                80.2  \n",
              "9232       1667.0       724.0            4.9               134.1  \n",
              "15285      1361.0       608.0            2.0               147.4  \n",
              "7025       2420.0       947.0            4.0               277.9  \n",
              "9612       1031.0       259.0            1.7                48.0  \n",
              "...           ...         ...            ...                 ...  \n",
              "10554      1402.0       471.0            2.0               134.0  \n",
              "13842      1189.0       573.0            3.9               301.1  \n",
              "4555       1377.0       395.0            6.6               350.4  \n",
              "1962       1034.0       407.0            6.4               295.1  \n",
              "14377      1173.0       472.0            4.2               266.4  \n",
              "\n",
              "[17000 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 1
        }
      ]
    },
    {
      "metadata": {
        "id": "I6kNgrwCO_ms",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Next, we'll set up our input function, and define the function for model training:"
      ]
    },
    {
      "metadata": {
        "id": "5RpTJER9XDub",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "e15e5d96-28a4-44d3-a4d7-32d480a6f2c5"
      },
      "cell_type": "code",
      "source": [
        "def my_input_fn(features, targets, batch_size=1, shuffle=True, num_epochs=None):\n",
        "    \"\"\"Trains a linear regression model of one feature.\n",
        "  \n",
        "    Args:\n",
        "      features: pandas DataFrame of features\n",
        "      targets: pandas DataFrame of targets\n",
        "      batch_size: Size of batches to be passed to the model\n",
        "      shuffle: True or False. Whether to shuffle the data.\n",
        "      num_epochs: Number of epochs for which data should be repeated. None = repeat indefinitely\n",
        "    Returns:\n",
        "      Tuple of (features, labels) for next data batch\n",
        "    \"\"\"\n",
        "    \n",
        "    # Convert pandas data into a dict of np arrays.\n",
        "    features = {key:np.array(value) for key,value in dict(features).items()}                                           \n",
        " \n",
        "    # Construct a dataset, and configure batching/repeating.\n",
        "    ds = Dataset.from_tensor_slices((features,targets)) # warning: 2GB limit\n",
        "    ds = ds.batch(batch_size).repeat(num_epochs)\n",
        "    \n",
        "    # Shuffle the data, if specified.\n",
        "    if shuffle:\n",
        "      ds = ds.shuffle(buffer_size=10000)\n",
        "    \n",
        "    # Return the next batch of data.\n",
        "    features, labels = ds.make_one_shot_iterator().get_next()\n",
        "    return features, labels"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "id": "VgQPftrpHbG3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "122d2a34-9e3c-4e66-f8ac-543ffcb1df57"
      },
      "cell_type": "code",
      "source": [
        "def train_model(learning_rate, steps, batch_size, input_feature):\n",
        "  \"\"\"Trains a linear regression model.\n",
        "  \n",
        "  Args:\n",
        "    learning_rate: A `float`, the learning rate.\n",
        "    steps: A non-zero `int`, the total number of training steps. A training step\n",
        "      consists of a forward and backward pass using a single batch.\n",
        "    batch_size: A non-zero `int`, the batch size.\n",
        "    input_feature: A `string` specifying a column from `california_housing_dataframe`\n",
        "      to use as input feature.\n",
        "      \n",
        "  Returns:\n",
        "    A Pandas `DataFrame` containing targets and the corresponding predictions done\n",
        "    after training the model.\n",
        "  \"\"\"\n",
        "  \n",
        "  periods = 10\n",
        "  steps_per_period = steps / periods\n",
        "\n",
        "  my_feature = input_feature\n",
        "  my_feature_data = california_housing_dataframe[[my_feature]].astype('float32')\n",
        "  my_label = \"median_house_value\"\n",
        "  targets = california_housing_dataframe[my_label].astype('float32')\n",
        "\n",
        "  # Create input functions.\n",
        "  training_input_fn = lambda: my_input_fn(my_feature_data, targets, batch_size=batch_size)\n",
        "  predict_training_input_fn = lambda: my_input_fn(my_feature_data, targets, num_epochs=1, shuffle=False)\n",
        "  \n",
        "  # Create feature columns.\n",
        "  feature_columns = [tf.feature_column.numeric_column(my_feature)]\n",
        "    \n",
        "  # Create a linear regressor object.\n",
        "  my_optimizer = tf.train.GradientDescentOptimizer(learning_rate=learning_rate)\n",
        "  my_optimizer = tf.contrib.estimator.clip_gradients_by_norm(my_optimizer, 5.0)\n",
        "  linear_regressor = tf.estimator.LinearRegressor(\n",
        "      feature_columns=feature_columns,\n",
        "      optimizer=my_optimizer\n",
        "  )\n",
        "\n",
        "  # Set up to plot the state of our model's line each period.\n",
        "  plt.figure(figsize=(15, 6))\n",
        "  plt.subplot(1, 2, 1)\n",
        "  plt.title(\"Learned Line by Period\")\n",
        "  plt.ylabel(my_label)\n",
        "  plt.xlabel(my_feature)\n",
        "  sample = california_housing_dataframe.sample(n=300)\n",
        "  plt.scatter(sample[my_feature], sample[my_label])\n",
        "  colors = [cm.coolwarm(x) for x in np.linspace(-1, 1, periods)]\n",
        "\n",
        "  # Train the model, but do so inside a loop so that we can periodically assess\n",
        "  # loss metrics.\n",
        "  print \"Training model...\"\n",
        "  print \"RMSE (on training data):\"\n",
        "  root_mean_squared_errors = []\n",
        "  for period in range (0, periods):\n",
        "    # Train the model, starting from the prior state.\n",
        "    linear_regressor.train(\n",
        "        input_fn=training_input_fn,\n",
        "        steps=steps_per_period,\n",
        "    )\n",
        "    # Take a break and compute predictions.\n",
        "    predictions = linear_regressor.predict(input_fn=predict_training_input_fn)\n",
        "    predictions = np.array([item['predictions'][0] for item in predictions])\n",
        "    \n",
        "    # Compute loss.\n",
        "    root_mean_squared_error = math.sqrt(\n",
        "      metrics.mean_squared_error(predictions, targets))\n",
        "    # Occasionally print the current loss.\n",
        "    print \"  period %02d : %0.2f\" % (period, root_mean_squared_error)\n",
        "    # Add the loss metrics from this period to our list.\n",
        "    root_mean_squared_errors.append(root_mean_squared_error)\n",
        "    # Finally, track the weights and biases over time.\n",
        "    # Apply some math to ensure that the data and line are plotted neatly.\n",
        "    y_extents = np.array([0, sample[my_label].max()])\n",
        "    \n",
        "    weight = linear_regressor.get_variable_value('linear/linear_model/%s/weights' % input_feature)[0]\n",
        "    bias = linear_regressor.get_variable_value('linear/linear_model/bias_weights')\n",
        "    \n",
        "    x_extents = (y_extents - bias) / weight\n",
        "    x_extents = np.maximum(np.minimum(x_extents,\n",
        "                                      sample[my_feature].max()),\n",
        "                           sample[my_feature].min())\n",
        "    y_extents = weight * x_extents + bias\n",
        "    plt.plot(x_extents, y_extents, color=colors[period]) \n",
        "  print \"Model training finished.\"\n",
        "\n",
        "  # Output a graph of loss metrics over periods.\n",
        "  plt.subplot(1, 2, 2)\n",
        "  plt.ylabel('RMSE')\n",
        "  plt.xlabel('Periods')\n",
        "  plt.title(\"Root Mean Squared Error vs. Periods\")\n",
        "  plt.tight_layout()\n",
        "  plt.plot(root_mean_squared_errors)\n",
        "\n",
        "  # Create a table with calibration data.\n",
        "  calibration_data = pd.DataFrame()\n",
        "  calibration_data[\"predictions\"] = pd.Series(predictions)\n",
        "  calibration_data[\"targets\"] = pd.Series(targets)\n",
        "  display.display(calibration_data.describe())\n",
        "\n",
        "  print \"Final RMSE (on training data): %0.2f\" % root_mean_squared_error\n",
        "  \n",
        "  return calibration_data"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "id": "FJ6xUNVRm-do",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Task 1: Try a Synthetic Feature\n",
        "\n",
        "Both the `total_rooms` and `population` features count totals for a given city block.\n",
        "\n",
        "But what if one city block were more densely populated than another? We can explore how block density relates to median house value by creating a synthetic feature that's a ratio of `total_rooms` and `population`.\n",
        "\n",
        "In the cell below, create a feature called `rooms_per_person`, and use that as the `input_feature` to `train_model()`.\n",
        "\n",
        "What's the best performance you can get with this single feature by tweaking the learning rate? (The better the performance, the better your regression line should fit the data, and the lower\n",
        "the final RMSE should be.)"
      ]
    },
    {
      "metadata": {
        "id": "isONN2XK32Wo",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**NOTE**: You may find it helpful to add a few code cells below so you can try out several different learning rates and compare the results. To add a new code cell, hover your cursor directly below the center of this cell, and click **CODE**."
      ]
    },
    {
      "metadata": {
        "id": "5ihcVutnnu1D",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 950
        },
        "outputId": "cdfa3945-6954-4d71-9d59-603c3f4f8ef8"
      },
      "cell_type": "code",
      "source": [
        "#\n",
        "# YOUR CODE HERE\n",
        "#\n",
        "\n",
        "california_housing_dataframe[\"rooms_per_person\"] =(california_housing_dataframe[\"total_rooms\"]/california_housing_dataframe[\"population\"])\n",
        "\n",
        "calibration_data = train_model(\n",
        "    learning_rate=0.0001,\n",
        "    steps=15,\n",
        "    batch_size=10,\n",
        "    input_feature=\"rooms_per_person\"\n",
        ")"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training model...\n",
            "RMSE (on training data):\n",
            "  period 00 : 237.54\n",
            "  period 01 : 237.54\n",
            "  period 02 : 237.54\n",
            "  period 03 : 237.54\n",
            "  period 04 : 237.53\n",
            "  period 05 : 237.53\n",
            "  period 06 : 237.53\n",
            "  period 07 : 237.53\n",
            "  period 08 : 237.53\n",
            "  period 09 : 237.53\n",
            "Model training finished.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "       predictions  targets\n",
              "count      17000.0  17000.0\n",
              "mean           0.0    207.3\n",
              "std            0.0    116.0\n",
              "min            0.0     15.0\n",
              "25%            0.0    119.4\n",
              "50%            0.0    180.4\n",
              "75%            0.0    265.0\n",
              "max            0.2    500.0"
            ],
            "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>predictions</th>\n",
              "      <th>targets</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>17000.0</td>\n",
              "      <td>17000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.0</td>\n",
              "      <td>207.3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.0</td>\n",
              "      <td>116.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.0</td>\n",
              "      <td>15.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.0</td>\n",
              "      <td>119.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.0</td>\n",
              "      <td>180.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.0</td>\n",
              "      <td>265.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>0.2</td>\n",
              "      <td>500.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Final RMSE (on training data): 237.53\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABCUAAAGkCAYAAAAG3J9IAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f837949a590>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "i5Ul3zf5QYvW",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Solution\n",
        "\n",
        "Click below for a solution."
      ]
    },
    {
      "metadata": {
        "id": "Leaz2oYMQcBf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "california_housing_dataframe[\"rooms_per_person\"] = (\n",
        "    california_housing_dataframe[\"total_rooms\"] / california_housing_dataframe[\"population\"])\n",
        "\n",
        "calibration_data = train_model(\n",
        "    learning_rate=0.05,\n",
        "    steps=500,\n",
        "    batch_size=5,\n",
        "    input_feature=\"rooms_per_person\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZjQrZ8mcHFiU",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Task 2: Identify Outliers\n",
        "\n",
        "We can visualize the performance of our model by creating a scatter plot of predictions vs. target values.  Ideally, these would lie on a perfectly correlated diagonal line.\n",
        "\n",
        "Use Pyplot's [`scatter()`](https://matplotlib.org/gallery/shapes_and_collections/scatter.html) to create a scatter plot of predictions vs. targets, using the rooms-per-person model you trained in Task 1.\n",
        "\n",
        "Do you see any oddities?  Trace these back to the source data by looking at the distribution of values in `rooms_per_person`."
      ]
    },
    {
      "metadata": {
        "id": "P0BDOec4HbG_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "outputId": "d84fc1b6-2c6a-4bd8-d599-973fa7f14a4c"
      },
      "cell_type": "code",
      "source": [
        "# YOUR CODE HERE\n",
        "\n",
        "plt.scatter(calibration_data[\"predictions\"], calibration_data[\"targets\"])"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7f8379564d90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAFlCAYAAACp5uxjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f838088e150>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "jByCP8hDRZmM",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Solution\n",
        "\n",
        "Click below for the solution."
      ]
    },
    {
      "metadata": {
        "id": "s0tiX2gdRe-S",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "outputId": "0b97f851-97b9-4728-b1c5-94cfb9d4da1f"
      },
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(15, 6))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.scatter(calibration_data[\"predictions\"], calibration_data[\"targets\"])"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7f83796f4a90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAFlCAYAAACp5uxjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f837b863f10>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "kMQD0Uq3RqTX",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "The calibration data shows most scatter points aligned to a line. The line is almost vertical, but we'll come back to that later. Right now let's focus on the ones that deviate from the line. We notice that they are relatively few in number.\n",
        "\n",
        "If we plot a histogram of `rooms_per_person`, we find that we have a few outliers in our input data:"
      ]
    },
    {
      "metadata": {
        "id": "POTM8C_ER1Oc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 347
        },
        "outputId": "a628c1c2-bcd6-4075-b406-c82c30ab9495"
      },
      "cell_type": "code",
      "source": [
        "plt.subplot(1, 2, 2)\n",
        "_ = california_housing_dataframe[\"rooms_per_person\"].hist()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPwAAAFKCAYAAAA5TzK7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfIzt3Nflkr4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f837955fbd0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "9l0KYpBQu8ed",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Task 3: Clip Outliers\n",
        "\n",
        "See if you can further improve the model fit by setting the outlier values of `rooms_per_person` to some reasonable minimum or maximum.\n",
        "\n",
        "For reference, here's a quick example of how to apply a function to a Pandas `Series`:\n",
        "\n",
        "    clipped_feature = my_dataframe[\"my_feature_name\"].apply(lambda x: max(x, 0))\n",
        "\n",
        "The above `clipped_feature` will have no values less than `0`."
      ]
    },
    {
      "metadata": {
        "id": "rGxjRoYlHbHC",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# YOUR CODE HERE"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "WvgxW0bUSC-c",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Solution\n",
        "\n",
        "Click below for the solution."
      ]
    },
    {
      "metadata": {
        "id": "8YGNjXPaSMPV",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "The histogram we created in Task 2 shows that the majority of values are less than `5`. Let's clip `rooms_per_person` to 5, and plot a histogram to double-check the results."
      ]
    },
    {
      "metadata": {
        "id": "9YyARz6gSR7Q",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 347
        },
        "outputId": "352f6d1a-9a39-40e0-8621-0bc6770d2d89"
      },
      "cell_type": "code",
      "source": [
        "california_housing_dataframe[\"rooms_per_person\"] = (\n",
        "    california_housing_dataframe[\"rooms_per_person\"]).apply(lambda x: min(x, 5))\n",
        "\n",
        "_ = california_housing_dataframe[\"rooms_per_person\"].hist()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAFKCAYAAADScRzUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQg0AgGGEGgAAw/4P1DJKJgyt6msAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f837c7d9c90>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "vO0e1p_aSgKA",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "To verify that clipping worked, let's train again and print the calibration data once more:"
      ]
    },
    {
      "metadata": {
        "id": "ZgSP2HKfSoOH",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "calibration_data = train_model(\n",
        "    learning_rate=0.05,\n",
        "    steps=500,\n",
        "    batch_size=5,\n",
        "    input_feature=\"rooms_per_person\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "gySE-UgfSony",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "_ = plt.scatter(calibration_data[\"predictions\"], calibration_data[\"targets\"])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "2T5RekGVbDIj",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}