imdbBinClass.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "imdbBinClass.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "8Q13iq-KaEVm",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from keras.datasets import imdb\n",
        "from keras import models\n",
        "from keras import layers\n",
        "from keras.utils import to_categorical\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from keras import regularizers"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "6oZSXGtgbG7v",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "(train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=10000)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0H8zp9s0MneN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "word_index = imdb.get_word_index()\n",
        "word_index = dict([(value, key) for (key,value) in word_index.items()])\n",
        "decoded_review = [word_index.get(i - 3, '?') for i in train_data[0]]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "NbSK8odIbEgn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7eceaa56-3c71-437f-8d92-0989a20dabf9"
      },
      "cell_type": "code",
      "source": [
        "len(train_data)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "25000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "metadata": {
        "id": "Tf-DPTyUMoEH",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def vectorize_sequences(sequences, dimension=10000):\n",
        "  results = np.zeros((len(sequences), dimension))\n",
        "  for i, sequence in enumerate(sequences):\n",
        "    results[i,sequence] = 1\n",
        "  return results\n",
        "\n",
        "x_train = vectorize_sequences(train_data)\n",
        "x_text = vectorize_sequences(test_data)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "34rf5cJVbhEr",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "844cd531-6b5b-452e-dd03-35c17d54b986"
      },
      "cell_type": "code",
      "source": [
        "type(x_train)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "numpy.ndarray"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "metadata": {
        "id": "JxRsHERiMU-Z",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "y_train = np.asarray(train_labels).astype('float32')\n",
        "y_test = np.asarray(test_labels).astype('float32')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "GAU06AXIvEXl",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = models.Sequential()\n",
        "model.add(layers.Dense(16,kernel_regularizer=regularizers.l2(0.001), activation='relu', input_shape=(10000,)))\n",
        "model.add(layers.Dropout(0.5))\n",
        "model.add(layers.Dense(16,kernel_regularizer=regularizers.l2(0.001), activation='relu'))\n",
        "model.add(layers.Dropout(0.5))\n",
        "model.add(layers.Dense(1, activation='sigmoid'))\n",
        "\n",
        "model.compile(optimizer='rmsprop',\n",
        "              loss='binary_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ong_X8epbVjY",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "x_val = x_train[:10000]\n",
        "partial_x_train = x_train[10000:]\n",
        "\n",
        "y_val = y_train[:10000]\n",
        "partial_y_train = y_train[10000:]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "W_iweCOGHQfy",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "20eb1ddc-b1c2-4aff-a5c4-2bc0819ca107"
      },
      "cell_type": "code",
      "source": [
        "type(y_val)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "numpy.ndarray"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "metadata": {
        "id": "1kbbzW0UbYX3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "b71437b4-ea75-4056-8bae-eac1718599b5"
      },
      "cell_type": "code",
      "source": [
        "history = model.fit(partial_x_train,partial_y_train, epochs=4, batch_size=512, validation_data=(x_val,y_val))"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 2s 127us/step - loss: 0.6708 - acc: 0.6359 - val_loss: 0.5758 - val_acc: 0.8317\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 109us/step - loss: 0.5643 - acc: 0.7464 - val_loss: 0.4700 - val_acc: 0.8715\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 109us/step - loss: 0.4864 - acc: 0.8061 - val_loss: 0.3965 - val_acc: 0.8807\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 108us/step - loss: 0.4289 - acc: 0.8461 - val_loss: 0.3520 - val_acc: 0.8847\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "YtLx1nVobcau",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "15f74ba2-4a96-4203-bda9-ae777caed865"
      },
      "cell_type": "code",
      "source": [
        "history_dict = history.history\n",
        "loss_values = history_dict['loss']\n",
        "val_loss_values = history_dict['val_loss']\n",
        "acc = history_dict['acc']\n",
        "epochs = range(1, len(acc) + 1)\n",
        "history_dict.keys()"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dict_keys(['val_loss', 'val_acc', 'loss', 'acc'])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "metadata": {
        "id": "vsiCeg7RbdhU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 376
        },
        "outputId": "35f8ba8f-201a-47ab-f0ce-20fb7adc56bb"
      },
      "cell_type": "code",
      "source": [
        "plt.figure()\n",
        "plt.plot(epochs, loss_values, 'bo', label='Training loss')\n",
        "plt.plot(epochs, val_loss_values, 'b', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nACFEHQBCiDoAhLgBOIpT71OG8TMAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc9990ea9e8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "fiUnVkMHLA_M",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 376
        },
        "outputId": "599571ea-77ea-4276-b120-d77c740a96e7"
      },
      "cell_type": "code",
      "source": [
        "acc_values = history_dict['acc']\n",
        "val_acc_values = history_dict['val_acc']\n",
        "plt.plot(epochs, acc, 'bo', label='Training acc')\n",
        "plt.plot(epochs, val_acc_values, 'b', label='Validation acc')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUKiLiIgYhEJdRETEIP4/DVZ0WPVWRQsAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc9a1d45ef0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "fRiWYP4gLIK9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7df0b945-b45f-43cd-edf0-5919b0c1d647"
      },
      "cell_type": "code",
      "source": [
        "results = model.evaluate(x_text, y_test)"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "25000/25000 [==============================] - 3s 113us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Lp1DF6srLxQ7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0a37072d-a878-4c9e-b543-e7466f35427f"
      },
      "cell_type": "code",
      "source": [
        "results"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.36303951608657836, 0.88084]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "metadata": {
        "id": "cy2G5CJ2MMEQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "c0f88d18-2fe4-4578-d924-69c58b26fd11"
      },
      "cell_type": "code",
      "source": [
        "model.predict(x_text)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.31602818],\n",
              "       [0.9717633 ],\n",
              "       [0.8934865 ],\n",
              "       ...,\n",
              "       [0.18426757],\n",
              "       [0.2419499 ],\n",
              "       [0.47101322]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "metadata": {
        "id": "55UMNgGMMXtH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "5da3a49d-d8e9-446d-b684-94793b946b9e"
      },
      "cell_type": "code",
      "source": [
        "np.random.randint(0, high=2, size=(4,4))"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1, 1, 0, 1],\n",
              "       [0, 0, 1, 0],\n",
              "       [1, 0, 1, 1],\n",
              "       [0, 1, 1, 1]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "metadata": {
        "id": "Rekon_RlEi1N",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}