Must use google colab. Please design a LSTM-based recurrent neural network using python to complete...

Question

Must use google colab. Please design a LSTM-based recurrent neural network using python to complete the classification task for IMDB movie review sentiment classification dataset. Please submit your codes together with analysis about integrating convolutional neural network to the LSTM-based recurrent neural network in this task. You may use the LSTM tutorial attachment as a guide to creating your solution. Please run the code and then create a pdf for submission.{   "nbformat": 4,   "nbformat_minor": 0,   "metadata": {     "colab": {       "name": "LSTM_Tutorial.ipynb",       "provenance": [],       "collapsed_sections": []     },     "kernelspec": {       "name": "python3",       "display_name": "Python 3"     },     "accelerator": "GPU"   },   "cells": [     {       "cell_type": "code",       "metadata": {         "id": "rcTzbmQBYQB6"       },       "source": [         "import tensorflow as tf
",         "from tensorflow.keras import datasets, layers, models, optimizers
",         "from tensorflow.keras.datasets import imdb
",         "from tensorflow.keras.models import Sequential
",         "from tensorflow.keras.optimizers import SGD, Adam
",         "from tensorflow.keras.layers import Input, LSTM, Dense, Embedding, Conv1D, MaxPooling1D
",         "from tensorflow.keras.preprocessing import sequence
",         "
",         "import matplotlib.pyplot as plt
",         "import numpy as np
",         "import pandas as pd
"       ],       "execution_count": 1,       "outputs": []     },     {       "cell_type": "code",       "metadata": {         "id": "1sIigNQMi8WL",         "colab": {           "base_uri": "https://localhost:8080/"         },         "outputId": "86c2aa2b-6030-463d-8f6a-58ec0b71671a"       },       "source": [         "# load the dataset but only keep the top n words, zero the rest
",         "# The IMDB dataset is a set of 50,000 highly polarized reviews from the Internet Movie Database. Each set contains an equal number (50%) of positive and negative reviews. They are split into 25000 reviews each for training and testing
",         "top_words = 5000
",         "(x_train, y_train), (x_test, y_test) = imdb.load_data(num_words=top_words)"       ],       "execution_count": 2,       "outputs": [         {           "output_type": "stream",           "text": [             ":6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",             "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:159: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",             "  x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
",             "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:160: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",             "  x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
"           ],           "name": "stderr"         }       ]     },     {       "cell_type": "code",       "metadata": {         "id": "ozhJjRoQrPbl"       },       "source": [         "# truncate and pad input sequences to ensure the vectors have the same length although the length of the individual content can be different
",         "max_review_length = 500
",         "x_train = sequence.pad_sequences(x_train, maxlen=max_review_length)
",         "x_test = sequence.pad_sequences(x_test, maxlen=max_review_length)"       ],       "execution_count": 3,       "outputs": []     },     {       "cell_type": "code",       "metadata": {         "id": "esSqg8N2sPSo",         "colab": {           "base_uri": "https://localhost:8080/"         },         "outputId": "603ee06b-6ca3-4328-e67b-378e4e35ef7b"       },       "source": [         "# Build LSTM
",         "# The first layer is the Embedded layer that uses the vectors with length of 32 to present each word
",         "# The second layer is the LSTM layer with 100 units
",         "# The output layer is Dense layer since this is a classification problem. Because this is a binary classification, we can use sigmoid instead of softmax
",         "model = Sequential()
",         "embedding_vecor_length = 32
",         "model.add(Embedding(top_words, embedding_vecor_length, input_length=max_review_length))
",         "model.add(LSTM(100))
",         "model.add(Dense(1, activation='sigmoid'))
",         "print(model.summary())"       ],       "execution_count": 4,       "outputs": [         {           "output_type": "stream",           "text": [             "Model: "sequential"
",             "_________________________________________________________________
",             "Layer (type)                 Output Shape              Param #   
",             "=================================================================
",             "embedding (Embedding)        (None, 500, 32)           160000    
",             "_________________________________________________________________
",             "lstm (LSTM)                  (None, 100)               53200     
",             "_________________________________________________________________
",             "dense (Dense)                (None, 1)                 101       
",             "=================================================================
",             "Total params: 213,301
",             "Trainable params: 213,301
",             "Non-trainable params: 0
",             "_________________________________________________________________
",             "None
"           ],           "name": "stdout"         }       ]     },     {       "cell_type": "code",       "metadata": {         "id": "GHIW82-pwXu4",         "colab": {           "base_uri": "https://localhost:8080/"         },         "outputId": "5a494da4-54b2-4ca9-ea93-b4f77f894604"       },       "source": [         "# Compile and train the learning model
",         "adam = optimizers.Adam(lr=0.001)
",         "
",         "model.compile(adam, loss='binary_crossentropy', metrics=['accuracy'])
",         "
",         "history = model.fit(x_train, y_train, validation_data=(x_test, y_test), epochs=3, batch_size=64)"       ],       "execution_count": 5,       "outputs": [         {           "output_type": "stream",           "text": [             "Epoch 1/3
",

Sandeep Kumar · Accepted Answer

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "name": "python394jvsc74a57bd081118431cc388d258ed977b65143603a98f8ad6ed776c173758a3af876bc6de9",
      "display_name": "Python 3.9.4 64-bit"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.9.4-final"
    },
    "colab": {
      "name": "IMDB movie review Modeling DNN CNN LSTM Keras.ipynb",
      "provenance": []
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "yJDfTCHvULWg"
      },
      "source": [
        "from keras.datasets import imdb
",
        "%matplotlib inline
",
        "import numpy as np
",
        "import pandas as pd
",
        "from matplotlib import cm
",
        "import matplotlib.pyplot as plt
",
        "import seaborn as sns
",
        "import os
",
        "import time"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0Lf43x1oULW7"
      },
      "source": [
        "from keras.preprocessing import sequence
",
        "from keras.models import Sequential
",
        "from keras.layers import Dense, Dropout, Activation
",
        "from keras.layers import Embedding
",
        "from keras.layers import Conv1D, GlobalMaxPooling1D
",
        "from keras.callbacks import EarlyStopping
",
        "from keras import models"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "9Z2RhfpPULW-",
        "outputId": "dffec856-2810-49de-b253-5c0d3bb524e6"
      },
      "source": [
        "
",
        "(X_train, y_train), (X_test, y_test) = imdb.load_data()
",
        "X = np.concatenate((X_train, X_test), axis=0)
",
        "y = np.concatenate((y_train, y_test), axis=0)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            ":6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:159: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "  x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
",
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:160: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "  x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "JJXh9NkDULW_",
        "outputId": "27ee5cf9-43e0-4501-acdb-532c3e635253"
      },
      "source": [
        "##training data shape review
",
        "print("Training data: ")
",
        "print(X.shape)
",
        "print(y.shape)
",
        "print("Classes: ")
",
        "print(np.unique(y))"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training data: 
",
            "(50000,)
",
            "(50000,)
",
            "Classes: 
",
            "[0 1]
"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "GnKYq1bVULXB",
        "outputId": "0adaa77d-9b67-4e58-ad7d-1f62da898acb"
      },
      "source": [
        "print("Number of words: ")
",
        "print(len(np.unique(np.hstack(X))))"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of words: 
",
            "88585
"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        },
        "id": "ffaOSrvEULXC",
        "outputId": "10a0455a-d45f-4a7b-8c0c-dc1456211e71"
      },
      "source": [
        "print("Review length: ")
",
        "result = [len(x) for x in X]
",
        "print("Mean %.2f words (%f)" % (np.mean(result), np.std(result)))
",
        "# plot review length
",
        "plt.boxplot(result)
",
        "plt.show()"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Review length: 
",
            "Mean 234.76 words (172.911495)
"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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
",
            "text/plain": [
              ""
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lRyNiW0sULXE"
      },
      "source": [
        "#####Analysis#####"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "6gnnvsvvULXF",
        "outputId": "ff1c4c17-c31d-44c1-c2e5-524a636ff310"
      },
      "source": [
        "(train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=5000)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            ":6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:159: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "  x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
",
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/datasets/imdb.py:160: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
",
            "  x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JBanO5PMULXG"
      },
      "source": [
        "def vectorize_sequences(sequences, dimension=5000):
",
        "    # Create an all-zero matrix of shape (len(sequences), dimension)
",
        "    results = np.zeros((len(sequences), dimension))
",
        "    for i, sequence in enumerate(sequences):
",
        "        results[i, sequence] = 1.  # set specific indices of results[i] to 1s
",
        "    return results
",
        "
",
        "# Our vectorized training data
",
        "x_train = vectorize_sequences(train_data)
",
        "# Our vectorized test data
",
        "x_test = vectorize_sequences(test_data)"
      ],
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m2UgUmlGULXI"
      },
      "source": [
        "# Our vectorized labels one-hot encoder
",
        "y_train = np.asarray(train_labels).astype('float32')
",
        "y_test = np.asarray(test_labels).astype('float32')"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "995zJLPtULXJ"
      },
      "source": [
        "from keras import layers
",
        "from keras import models
",
        "
",
        "model = models.Sequential()
",
        "model.add(layers.Dense(32, activation='relu', input_shape=(5000,)))
",
        "model.add(layers.Dense(32, activation='relu',))
",
        "model.add(layers.Dense(1, activation='sigmoid'))"
      ],
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u6V3sbfeULXK"
      },
      "source": [
        "#Set validation set aside
",
        "
",
        "x_val = x_train[:10000]
",
        "partial_x_train = x_train[10000:]
",
        "
",
        "y_val = y_train[:10000]
",
        "partial_y_train = y_train[10000:]"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gUl463vOULXL"
      },
      "source": [
        "model.compile(optimizer='adam',
",
        "              loss='binary_crossentropy',
",
        "              metrics=['acc'])"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "Q41E98NWULXM",
        "outputId": "f5a27d70-0896-4f4a-cfe3-54d3642afb56"
      },
      "source": [
        "start_time_m1 = time.time()
",
        "history = model.fit(partial_x_train,
",
        "                    partial_y_train,
",
        "                    epochs=20,
",
        "                    batch_size=512,
",
        "                    validation_data=(x_val, y_val))
",
        "total_time_m1 = time.time() - start_time_m1
",
        "
",
        "print("The Dense Convolutional Neural Network 1 layer took %.4f seconds to train." % (total_time_m1))"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20
",
            "30/30 [==============================] - 1s 34ms/step - loss: 0.6197 - acc: 0.6813 - val_loss: 0.3706 - val_acc: 0.8568
",
            "Epoch 2/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.3011 - acc: 0.8871 - val_loss: 0.2959 - val_acc: 0.8813
",
            "Epoch 3/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.2059 - acc: 0.9258 - val_loss: 0.3023 - val_acc: 0.8825
",
            "Epoch 4/20
",
            "30/30 [==============================] - 1s 44ms/step - loss: 0.1605 - acc: 0.9426 - val_loss: 0.3152 - val_acc: 0.8759
",
            "Epoch 5/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.1310 - acc: 0.9543 - val_loss: 0.3386 - val_acc: 0.8757
",
            "Epoch 6/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.1130 - acc: 0.9656 - val_loss: 0.3741 - val_acc: 0.8709
",
            "Epoch 7/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0872 - acc: 0.9734 - val_loss: 0.3975 - val_acc: 0.8658
",
            "Epoch 8/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.0708 - acc: 0.9801 - val_loss: 0.4328 - val_acc: 0.8649
",
            "Epoch 9/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.0507 - acc: 0.9895 - val_loss: 0.4745 - val_acc: 0.8625
",
            "Epoch 10/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0345 - acc: 0.9959 - val_loss: 0.5124 - val_acc: 0.8636
",
            "Epoch 11/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0260 - acc: 0.9973 - val_loss: 0.5605 - val_acc: 0.8575
",
            "Epoch 12/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0199 - acc: 0.9990 - val_loss: 0.5915 - val_acc: 0.8603
",
            "Epoch 13/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.0136 - acc: 0.9994 - val_loss: 0.6290 - val_acc: 0.8579
",
            "Epoch 14/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0100 - acc: 0.9994 - val_loss: 0.6548 - val_acc: 0.8587
",
            "Epoch 15/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0074 - acc: 0.9999 - val_loss: 0.6876 - val_acc: 0.8575
",
            "Epoch 16/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0058 - acc: 1.0000 - val_loss: 0.7081 - val_acc: 0.8595
",
            "Epoch 17/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0044 - acc: 0.9999 - val_loss: 0.7309 - val_acc: 0.8581
",
            "Epoch 18/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0037 - acc: 1.0000 - val_loss: 0.7539 - val_acc: 0.8577
",
            "Epoch 19/20
",
            "30/30 [==============================] - 1s 24ms/step - loss: 0.0030 - acc: 1.0000 - val_loss: 0.7717 - val_acc: 0.8574
",
            "Epoch 20/20
",
            "30/30 [==============================] - 1s 25ms/step - loss: 0.0026 - acc: 1.0000 - val_loss: 0.7911 - val_acc: 0.8573
",
            "The Dense Convolutional Neural Network 1 layer took 16.1613 seconds to train.
"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "t5teUXrBULXN",
        "outputId": "7d47241e-9c94-410d-e275-e13c643ef5a4"
      },
      "source": [
        "history_dict = history.history
",
        "history_dict.keys()"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dict_keys(['loss', 'acc', 'val_loss', 'val_acc'])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "DU2YhCTQULXO",
        "outputId": "c6029e3d-6e18-4dd3-cac5-c41a97376f9c"
      },
      "source": [
        "import matplotlib.pyplot as plt
",
        "%matplotlib inline
",
        "
",
        "acc = history.history['acc']
",
        "val_acc = history.history['val_acc']
",
        "loss = history.history['loss']
",
        "val_loss = history.history['val_loss']
",
        "
",
        "epochs = range(1, len(acc) + 1)
",
        "
",
        "# "bo" is for "blue dot"
",
        "plt.plot(epochs, loss, 'bo', label='Training loss')
",
        "# b is for "solid blue line"
",
        "plt.plot(epochs, val_loss, 'b', label='Validation loss')
",
        "plt.title('Training and validation loss')
",
        "plt.xlabel('Epochs')
",
        "plt.ylabel('Loss')
",
        "plt.legend()
",
        "
",
        "plt.show()"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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
",
            "text/plain": [
              ""
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "ANUlU5ttULXQ",
        "outputId": "33509dc2-ab45-4916-e33f-c2fa4904bccd"
      },
      "source": [
        "plt.clf()   # clear figure
",
        "acc_values = history_dict['acc']
",
        "val_acc_values = history_dict['val_acc']
",
        "
",
        "plt.plot(epochs, acc, 'bo', label='Training acc')
",
        "plt.plot(epochs, val_acc, 'b', label='Validation acc')
",
        "plt.title('Training and validation accuracy')
",
        "plt.xlabel('Epochs')
",
        "plt.ylabel('Loss')
",
        "plt.legend()
",
        "
",
        "plt.show()"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png":

{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "colab": { "name": "LSTM_Tutorial.ipynb", "provenance": [], "collapsed_sections": [] }, "kernelspec": { "name": "python3", "display_name": "Python...

Answer To: { "nbformat": 4, "nbformat_minor": 0, "metadata": { "colab": { "name": "LSTM_Tutorial.ipynb",...

Answer To This Question Is Available To Download

Related Questions & Answers

Submit New Assignment