Neural_Clusters.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Neural Clusters.ipynb ",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/astrodatos/proyecto_neuralclusters/blob/master/Neural_Clusters.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ILrFj60aE-XA",
        "colab_type": "text"
      },
      "source": [
        "Se predecirá la temperatura efectiva de una estrella a través de sus magnitudes absolutas en cinco bandas fotométricas (problema de regresión). Se ha utilizado la base de datos del SDSS, consultando el mismo a través de astroquery."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_3BVRc6M-Ci8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!pip install --pre --quiet astroquery"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t5_yOjQBE4gz",
        "colab_type": "code",
        "outputId": "9db3879e-a8e6-41de-caee-2f545c695535",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "from __future__ import absolute_import, division, print_function, unicode_literals\n",
        "\n",
        "from time import time\n",
        "from astroquery.sdss import SDSS\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers\n",
        "\n",
        "\n",
        "print(tf.__version__)"
      ],
      "execution_count": 185,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.14.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hnipyp1PEwNi",
        "colab_type": "text"
      },
      "source": [
        "## Datos del SDSS\n",
        "\n",
        "Disponible en enlace\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KKOzVSMUrt1F",
        "colab_type": "text"
      },
      "source": [
        "###Consulta con Astroquery###"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l3cvu2Kj-iOB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "t_ini = time()\n",
        "\n",
        "data = SDSS.query_sql(\"\"\"SELECT TOP 100000\n",
        "                      p.fiberMag_u, p.fiberMag_g, p.fiberMag_r, p.fiberMag_i, \n",
        "                      p.fiberMag_z, s.elodieTEff\n",
        "                   \n",
        "                      \n",
        "                      FROM PhotoObjAll AS p JOIN specObjAll s ON s.bestobjid = p.objid\n",
        "                      \n",
        "                      WHERE p.mode = 1 AND s.sciencePrimary = 1 AND p.clean = 1 AND s.class = 'STAR'\n",
        "                      \n",
        "                      \"\"\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aYU1l2WHCUJI",
        "colab_type": "code",
        "outputId": "d858948c-dbc2-4096-f362-10e97a3cb31f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "datapd=data.to_pandas()\n",
        "datapd = datapd[(datapd != -9999).dropna()]\n",
        "datapd = datapd.dropna()\n",
        "print(datapd)"
      ],
      "execution_count": 198,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "       fiberMag_u  fiberMag_g  fiberMag_r  fiberMag_i  fiberMag_z  elodieTEff\n",
            "0        20.48503    18.48349    17.61969    17.21026    16.96410        4457\n",
            "1        18.33818    16.83223    16.19494    15.89421    15.72003        6155\n",
            "2        22.81749    21.94995    21.65532    21.49359    21.26813        5063\n",
            "3        21.87878    20.64922    20.36301    20.14630    19.93022        3705\n",
            "4        18.94203    18.00459    17.70750    17.59729    17.55546        3705\n",
            "5        18.20357    17.16672    17.05190    17.03187    17.02803        7500\n",
            "6        18.49530    17.25698    16.72305    16.51210    16.37836        5320\n",
            "7        16.81897    15.65400    15.16061    14.93756    14.83189        6000\n",
            "8        21.31538    18.69088    17.46540    16.98676    16.69788        4246\n",
            "9        20.26561    18.99056    18.46834    18.24545    18.12724        5873\n",
            "10       18.50102    16.36112    15.55532    15.26001    15.08880        5211\n",
            "11       25.68861    23.84693    22.21437    19.36426    17.64195        3717\n",
            "12       20.29619    18.89449    18.36372    18.09634    18.01541        5126\n",
            "13       20.22777    18.95881    18.44628    18.20092    18.08637        5792\n",
            "14       21.72731    19.19207    18.00698    17.50854    17.16981        4382\n",
            "15       24.18400    21.89584    20.49632    19.87902    19.09670        4949\n",
            "16       22.46864    21.50991    21.26591    21.16738    20.94329        9899\n",
            "17       19.25175    18.28875    18.17356    18.16572    18.17026        7000\n",
            "18       22.57421    21.80745    21.52110    21.51602    21.65648        4949\n",
            "19       24.21549    22.36420    22.04453    21.95734    21.47890        4949\n",
            "20       23.80392    22.26748    21.96685    21.66443    21.47087        4949\n",
            "21       22.82026    21.79802    21.72718    21.84858    21.46375        9899\n",
            "22       19.66879    18.72650    18.45093    18.34573    18.29824        3705\n",
            "23       25.00096    21.78606    20.64980    19.74212    19.15400        6163\n",
            "24       23.81696    22.41822    22.21108    22.31389    21.76348        5063\n",
            "25       22.04448    21.05343    20.87274    20.72685    20.40903        9899\n",
            "26       22.73199    21.32691    21.08852    20.99208    20.84533        3922\n",
            "27       21.89707    20.78792    20.61325    20.43084    20.41354        3705\n",
            "28       20.30603    19.36755    19.10143    18.99240    18.95734        3705\n",
            "29       21.54193    20.74511    20.35727    20.20912    19.84116        3900\n",
            "...           ...         ...         ...         ...         ...         ...\n",
            "99970    18.24890    17.35754    17.10670    17.02684    17.04664        6500\n",
            "99971    22.52203    19.89778    18.63595    18.86478    17.82514        3962\n",
            "99972    24.33316    22.51020    20.74672    18.64614    17.46614        3717\n",
            "99973    20.02781    18.72981    18.12391    17.89430    17.79525        5506\n",
            "99974    20.37206    18.96517    18.30838    18.08644    17.94760        5320\n",
            "99975    17.33471    16.34830    16.02086    15.88274    15.87687        6000\n",
            "99976    20.02494    18.29349    17.51785    17.24917    17.09679        4955\n",
            "99977    20.87655    18.78075    17.87547    17.48512    17.24702        4576\n",
            "99978    19.96440    18.26785    17.53427    17.22113    17.03591        5059\n",
            "99979    19.58145    18.09097    17.42179    17.16216    17.01911        5221\n",
            "99980    24.16771    24.46601    22.06507    20.39878    19.30951        4662\n",
            "99981    18.22255    16.36765    15.57151    15.25727    15.11043        4861\n",
            "99982    19.36206    17.66047    16.91505    16.61927    16.45943        5126\n",
            "99983    22.50180    21.29331    20.76824    20.59281    20.45070        6537\n",
            "99984    17.40181    16.39192    16.05211    15.92942    15.90278        6000\n",
            "99985    24.56870    24.44926    22.82210    21.24784    20.20386        4949\n",
            "99986    16.71023    15.45073    15.54114    15.62904    15.69266        7852\n",
            "99987    18.64260    16.57823    15.80880    15.54671    15.65177        4775\n",
            "99988    24.28972    23.82435    22.45616    20.92050    20.03629        6180\n",
            "99989    20.09649    20.28168    20.63116    20.80154    21.14259        9645\n",
            "99990    20.78385    18.64972    17.78449    17.48650    17.26370        4775\n",
            "99991    18.10546    16.35680    15.57919    15.27235    15.09041        4576\n",
            "99992    22.43680    20.50608    19.18018    18.53210    18.31663        3858\n",
            "99993    20.54170    19.52251    19.59503    19.48928    19.50327        7500\n",
            "99994    19.55521    18.56286    18.29286    18.28856    18.22897        3705\n",
            "99995    20.58042    18.39064    17.46416    17.15892    16.96882        4727\n",
            "99996    19.88992    17.79500    16.93505    16.66534    16.49487        4775\n",
            "99997    24.63640    22.79966    21.40135    20.25105    19.64192        4949\n",
            "99998    22.29836    21.30570    21.09404    20.99316    20.80540        3922\n",
            "99999    21.94150    20.43873    19.71582    19.39308    19.20578        5047\n",
            "\n",
            "[99986 rows x 6 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RbqnaZgMEOdV",
        "colab_type": "text"
      },
      "source": [
        "### Separación entre datos de entrenamiento y datos de prueba "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ytPIaTCAEMqe",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_datapd = datapd.sample(frac=0.8,random_state=0)\n",
        "test_datapd = datapd.drop(train_datapd.index)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "un9FMCKeEKJk",
        "colab_type": "text"
      },
      "source": [
        "### Chequeo a los datos"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hJMy5AApEEMv",
        "colab_type": "text"
      },
      "source": [
        "Estadísticas:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IXMZjr7HEBTf",
        "colab_type": "code",
        "outputId": "1878b5dd-9274-43b8-abb9-9410d2b4ba8f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "source": [
        "train_stats = train_datapd.describe()\n",
        "train_stats.pop(\"elodieTEff\")\n",
        "train_stats = train_stats.transpose()\n",
        "train_stats\n"
      ],
      "execution_count": 200,
      "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>count</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>min</th>\n",
              "      <th>25%</th>\n",
              "      <th>50%</th>\n",
              "      <th>75%</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>fiberMag_u</th>\n",
              "      <td>79989.0</td>\n",
              "      <td>20.982482</td>\n",
              "      <td>2.117595</td>\n",
              "      <td>14.01740</td>\n",
              "      <td>19.41306</td>\n",
              "      <td>20.80950</td>\n",
              "      <td>22.61272</td>\n",
              "      <td>29.04885</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fiberMag_g</th>\n",
              "      <td>79989.0</td>\n",
              "      <td>19.600113</td>\n",
              "      <td>1.998392</td>\n",
              "      <td>13.48535</td>\n",
              "      <td>18.12156</td>\n",
              "      <td>19.45434</td>\n",
              "      <td>21.28307</td>\n",
              "      <td>29.52714</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fiberMag_r</th>\n",
              "      <td>79989.0</td>\n",
              "      <td>19.002542</td>\n",
              "      <td>1.891515</td>\n",
              "      <td>12.84478</td>\n",
              "      <td>17.54710</td>\n",
              "      <td>18.96646</td>\n",
              "      <td>20.64430</td>\n",
              "      <td>24.66934</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fiberMag_i</th>\n",
              "      <td>79989.0</td>\n",
              "      <td>18.684114</td>\n",
              "      <td>1.822551</td>\n",
              "      <td>12.73436</td>\n",
              "      <td>17.28252</td>\n",
              "      <td>18.68764</td>\n",
              "      <td>20.17181</td>\n",
              "      <td>27.58280</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fiberMag_z</th>\n",
              "      <td>79989.0</td>\n",
              "      <td>18.497855</td>\n",
              "      <td>1.802802</td>\n",
              "      <td>12.13546</td>\n",
              "      <td>17.12093</td>\n",
              "      <td>18.46012</td>\n",
              "      <td>19.92276</td>\n",
              "      <td>24.00791</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              count       mean       std  ...       50%       75%       max\n",
              "fiberMag_u  79989.0  20.982482  2.117595  ...  20.80950  22.61272  29.04885\n",
              "fiberMag_g  79989.0  19.600113  1.998392  ...  19.45434  21.28307  29.52714\n",
              "fiberMag_r  79989.0  19.002542  1.891515  ...  18.96646  20.64430  24.66934\n",
              "fiberMag_i  79989.0  18.684114  1.822551  ...  18.68764  20.17181  27.58280\n",
              "fiberMag_z  79989.0  18.497855  1.802802  ...  18.46012  19.92276  24.00791\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 200
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MAH7deWED8tk",
        "colab_type": "text"
      },
      "source": [
        "### Separacaión entre atributos y etiquetas\n",
        "\n",
        "En este caso, la temperatura es la etiqueta, mientras que los atributos son las magnitudes absolutas en cinco bandas fotométricas."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1PsmBwsvD55i",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_labels = train_datapd.pop('elodieTEff')\n",
        "test_labels = test_datapd.pop('elodieTEff')\n",
        "\n",
        "train_feat = train_datapd\n",
        "test_feat = test_datapd\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q2fH1vq7D1Tw",
        "colab_type": "text"
      },
      "source": [
        "### Normalización de los datos\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lgbVr0RWSsom",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def norm(x):\n",
        "  r=(x-x.mean())/x.std()\n",
        "  return r"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mWW9QfEXDxnA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "norm_train_labels = norm(train_labels)\n",
        "norm_train_feat = norm(train_feat)\n",
        "\n",
        "norm_test_feat = norm(test_feat)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Di-_dkBrDrul",
        "colab_type": "text"
      },
      "source": [
        "Se utilizarán los datos normalizados para entrenar el modelo.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p9KUKRQFDlPn",
        "colab_type": "text"
      },
      "source": [
        "## El modelo"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fj8zlMCwvSKL",
        "colab_type": "text"
      },
      "source": [
        "###Estructura de la red neuronal"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Asg0O1TLDjAy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def build_model():\n",
        "  model = keras.Sequential([\n",
        "    layers.Dense(5, activation=tf.nn.tanh, input_shape=[len(train_datapd.keys())]),\n",
        "    #layers.Dense(1, activation=tf.nn.tanh),\n",
        "    #layers.Dense(4, activation=tf.nn.relu),\n",
        "    layers.Dense(1)\n",
        "  ])\n",
        "\n",
        "  optimizer = tf.keras.optimizers.RMSprop(0.001)\n",
        "\n",
        "  model.compile(loss='mean_squared_error',\n",
        "                optimizer=optimizer,\n",
        "                metrics=['mean_absolute_error', 'mean_squared_error'])\n",
        "  return model\n",
        "\n",
        "model = build_model()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fRs-X--DeKcQ",
        "colab_type": "text"
      },
      "source": [
        "###Desnormalización"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2aaIKxJbeJZA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def desnorm(dato, media, std):\n",
        "  datonuevo = dato*std + media\n",
        "  return datonuevo"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-LaH9J7aDb1B",
        "colab_type": "text"
      },
      "source": [
        "### Inspección del modelo"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eMeilODVDZ8L",
        "colab_type": "code",
        "outputId": "c4ee7f7c-3116-4650-b4cd-28ddc60c1dc7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 221
        }
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 206,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_11\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_33 (Dense)             (None, 5)                 30        \n",
            "_________________________________________________________________\n",
            "dense_34 (Dense)             (None, 1)                 6         \n",
            "=================================================================\n",
            "Total params: 36\n",
            "Trainable params: 36\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "elZqEAdBDKF8",
        "colab_type": "text"
      },
      "source": [
        "### Entrenamiento del modelo"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bW0KSl6iDHb0",
        "colab_type": "code",
        "outputId": "15764286-f869-4be3-e643-d2d5e43a2499",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "source": [
        "# Imprime un punto por cada epoch completado\n",
        "class PrintDot(keras.callbacks.Callback):\n",
        "  def on_epoch_end(self, epoch, logs):\n",
        "    if epoch % 100 == 0: print('')\n",
        "    print('.', end='')\n",
        "\n",
        "EPOCHS = 1000\n",
        "\n",
        "history = model.fit(\n",
        "  norm_train_feat, norm_train_labels,\n",
        "  epochs=EPOCHS, validation_split = 0.2, verbose=0,\n",
        "  callbacks=[PrintDot()])"
      ],
      "execution_count": 207,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "...................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "...................................................................................................."
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iRMaIjLWDCKA",
        "colab_type": "text"
      },
      "source": [
        "El progreso de entrenamiento del sistema está en el objeto `history`."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UHXhcQ-sDAN7",
        "colab_type": "code",
        "outputId": "a498cc46-0835-48dd-a0d9-f4e9d931ee47",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 566
        }
      },
      "source": [
        "hist = pd.DataFrame(history.history)\n",
        "hist['epoch'] = history.epoch\n",
        "hist.tail()\n",
        "\n",
        "print(history)\n",
        "\n",
        "def plot_history(history):\n",
        "  hist = pd.DataFrame(history.history)\n",
        "  hist['epoch'] = history.epoch\n",
        "\n",
        "  plt.figure()\n",
        "  plt.xlabel('Epoch')\n",
        "  plt.ylabel('Mean Abs Error [K]')\n",
        "  plt.plot(hist['epoch'], hist['mean_absolute_error'],\n",
        "           label='Train Error')\n",
        "  plt.plot(hist['epoch'], hist['val_mean_absolute_error'],\n",
        "           label = 'Val Error')\n",
        "  #plt.ylim([0,5])\n",
        "  plt.legend()\n",
        "\n",
        "  plt.figure()\n",
        "  plt.xlabel('Epoch')\n",
        "  plt.ylabel('Mean Square Error [$K^2$]')\n",
        "  plt.plot(hist['epoch'], hist['mean_squared_error'],\n",
        "           label='Train Error')\n",
        "  plt.plot(hist['epoch'], hist['val_mean_squared_error'],\n",
        "           label = 'Val Error')\n",
        "  #plt.ylim([0,20])\n",
        "  plt.legend()\n",
        "  plt.show()\n",
        "\n",
        "\n",
        "plot_history(history)"
      ],
      "execution_count": 208,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<tensorflow.python.keras.callbacks.History object at 0x7f3f4d64fc18>\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEKCAYAAADjDHn2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhWEYhpEVUxaGYRhGVkxZGIZhGFn5/xhJWGnxhUeWAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEKCAYAAADTgGjXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkggAm0QA2CQCwBYW3i+chzPB1QAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9OiJUrMvzz4k",
        "colab_type": "text"
      },
      "source": [
        "###Tiempo de Ejecución"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MBV0HbkYzzFa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "t_fin = time()\n",
        "t_ejecucion = t_fin - t_ini"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T3FTCj6gChEg",
        "colab_type": "text"
      },
      "source": [
        "### Predicción con los datos de prueba"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "go8xaaLkCdxN",
        "colab_type": "code",
        "outputId": "a643697e-86af-45d8-be63-875dcafee85e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 351
        }
      },
      "source": [
        "test_predictions = model.predict(norm_test_feat).flatten()\n",
        "\n",
        "test_predictions = desnorm(test_predictions, train_labels.mean(), train_labels.std())\n",
        "m1 = min(test_labels)\n",
        "M1 = max(test_labels)\n",
        "\n",
        "m2 = min(test_predictions)\n",
        "M2 = max(test_predictions)\n",
        "\n",
        "m = min([m1, m2])\n",
        "M = max([M1, M2])\n",
        "\n",
        "plt.scatter(test_labels, test_predictions,s=10, alpha = 0.003,color='red')\n",
        "plt.plot([m , M], [m, M])\n",
        "plt.xlabel('True Values [K]')\n",
        "plt.ylabel('Predictions [K]')\n",
        "plt.axis('equal')\n",
        "plt.axis('square')\n",
        "#plt.xlim([0,plt.xlim()[1]])\n",
        "#plt.ylim([0,100])\n",
        "#plt.ylim([0,plt.ylim()[1]])\n",
        "#_ = plt.plot([-100, 100], [-100, 100])"
      ],
      "execution_count": 223,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(19.266658399405628,\n",
              " 14227.852196232661,\n",
              " 19.259259340983363,\n",
              " 14227.844797174239)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 223
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAR8AAAEKCAYAAAArTFFlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ntmmeVgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bBQjihVVCWkc",
        "colab_type": "text"
      },
      "source": [
        "Notemos que nuestro modelo se debe acercar a la función de  identidad. Ahora veamos la distribución del error absoluto:\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m5gflQ0P81Ml",
        "colab_type": "text"
      },
      "source": [
        "### Errores"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o6Re1hClCN37",
        "colab_type": "code",
        "outputId": "5c8dfeb6-70ca-4dc6-f3fa-2d74bdcea065",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        }
      },
      "source": [
        "import numpy as np\n",
        "error = test_predictions - test_labels\n",
        "error2 = error**2\n",
        "\n",
        "plt.hist(error, bins = 30)\n",
        "plt.xlabel(\"Prediction Error [K]\")\n",
        "_ = plt.ylabel(\"Count\")\n",
        "\n",
        "MeanAbsError = np.abs(error).mean()\n",
        "MeanSQError = error2.mean()\n",
        "\n",
        "print('Tiempo de Ejecución:', t_ejecucion)\n",
        "print('Error Absoluto Medio:', MeanAbsError)\n",
        "print('Error Cuadrático Medio:', MeanSQError)"
      ],
      "execution_count": 211,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tiempo de Ejecución: 2846.7566928863525\n",
            "Error Absoluto Medio: 935.1488647460938\n",
            "Error Cuadrático Medio: 1763192.5\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEKCAYAAAAFJbKyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}