{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "accelerator": "GPU", "colab": { "name": "Course 3 - Week 2 - Lesson 2.ipynb", "provenance": [], "collapsed_sections": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "name": "python3" } }, "cells": [ { "cell_type": "code", "metadata": { "id": "zX4Kg8DUTKWO" }, "source": [ "#@title 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": null, "outputs": [] }, { "cell_type": "markdown", "metadata": { "id": "view-in-github" }, "source": [ "\"Open" ] }, { "cell_type": "code", "metadata": { "id": "4gs9htvM7n_x" }, "source": [ "# Run this to ensure TensorFlow 2.x is used\n", "try:\n", " # %tensorflow_version only exists in Colab.\n", " %tensorflow_version 2.x\n", "except Exception:\n", " pass" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "XYYDvoskkE61" }, "source": [ "import json\n", "import tensorflow as tf\n", "\n", "from tensorflow.keras.preprocessing.text import Tokenizer\n", "from tensorflow.keras.preprocessing.sequence import pad_sequences" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "0eJSTTYnkJQd" }, "source": [ "vocab_size = 10000\n", "embedding_dim = 16\n", "max_length = 100\n", "trunc_type='post'\n", "padding_type='post'\n", "oov_tok = \"\"\n", "training_size = 20000\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "BQVuQrZNkPn9" }, "source": [ "!wget --no-check-certificate \\\n", " https://storage.googleapis.com/laurencemoroney-blog.appspot.com/sarcasm.json \\\n", " -O /tmp/sarcasm.json\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "oaLaaqhNkUPd" }, "source": [ "with open(\"/tmp/sarcasm.json\", 'r') as f:\n", " datastore = json.load(f)\n", "\n", "sentences = []\n", "labels = []\n", "\n", "for item in datastore:\n", " sentences.append(item['headline'])\n", " labels.append(item['is_sarcastic'])" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "S1sD-7v0kYWk" }, "source": [ "training_sentences = sentences[0:training_size]\n", "testing_sentences = sentences[training_size:]\n", "training_labels = labels[0:training_size]\n", "testing_labels = labels[training_size:]" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "3u8UB0MCkZ5N" }, "source": [ "tokenizer = Tokenizer(num_words=vocab_size, oov_token=oov_tok)\n", "tokenizer.fit_on_texts(training_sentences)\n", "\n", "word_index = tokenizer.word_index\n", "\n", "training_sequences = tokenizer.texts_to_sequences(training_sentences)\n", "training_padded = pad_sequences(training_sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)\n", "\n", "testing_sequences = tokenizer.texts_to_sequences(testing_sentences)\n", "testing_padded = pad_sequences(testing_sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "GrAlWBKf99Ya" }, "source": [ "# Need this block to get it to work with TensorFlow 2.x\n", "import numpy as np\n", "training_padded = np.array(training_padded)\n", "training_labels = np.array(training_labels)\n", "testing_padded = np.array(testing_padded)\n", "testing_labels = np.array(testing_labels)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "FufaT4vlkiDE" }, "source": [ "model = tf.keras.Sequential([\n", " tf.keras.layers.Embedding(vocab_size, embedding_dim, input_length=max_length),\n", " tf.keras.layers.GlobalAveragePooling1D(),\n", " tf.keras.layers.Dense(24, activation='relu'),\n", " tf.keras.layers.Dense(1, activation='sigmoid')\n", "])\n", "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "XfDt1hmYkiys" }, "source": [ "model.summary()\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "2DTKQFf1kkyc" }, "source": [ "num_epochs = 30\n", "history = model.fit(training_padded, training_labels, epochs=num_epochs, validation_data=(testing_padded, testing_labels), verbose=2)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "2HYfBKXjkmU8" }, "source": [ "import matplotlib.pyplot as plt\n", "\n", "\n", "def plot_graphs(history, string):\n", " plt.plot(history.history[string])\n", " plt.plot(history.history['val_'+string])\n", " plt.xlabel(\"Epochs\")\n", " plt.ylabel(string)\n", " plt.legend([string, 'val_'+string])\n", " plt.show()\n", " \n", "plot_graphs(history, \"accuracy\")\n", "plot_graphs(history, \"loss\")" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "7SBdAZAenvzL" }, "source": [ "reverse_word_index = dict([(value, key) for (key, value) in word_index.items()])\n", "\n", "def decode_sentence(text):\n", " return ' '.join([reverse_word_index.get(i, '?') for i in text])\n", "\n", "print(decode_sentence(training_padded[0]))\n", "print(training_sentences[2])\n", "print(labels[2])" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "c9MqihtEkzQ9" }, "source": [ "e = model.layers[0]\n", "weights = e.get_weights()[0]\n", "print(weights.shape) # shape: (vocab_size, embedding_dim)\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "LoBXVffknldU" }, "source": [ "import io\n", "\n", "out_v = io.open('vecs.tsv', 'w', encoding='utf-8')\n", "out_m = io.open('meta.tsv', 'w', encoding='utf-8')\n", "for word_num in range(1, vocab_size):\n", " word = reverse_word_index[word_num]\n", " embeddings = weights[word_num]\n", " out_m.write(word + \"\\n\")\n", " out_v.write('\\t'.join([str(x) for x in embeddings]) + \"\\n\")\n", "out_v.close()\n", "out_m.close()" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "U4eZ5HtVnnEE" }, "source": [ "try:\n", " from google.colab import files\n", "except ImportError:\n", " pass\n", "else:\n", " files.download('vecs.tsv')\n", " files.download('meta.tsv')" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "cG8-ArY-qDcz" }, "source": [ "sentence = [\"granny starting to fear spiders in the garden might be real\", \"game of thrones season finale showing this sunday night\"]\n", "sequences = tokenizer.texts_to_sequences(sentence)\n", "padded = pad_sequences(sequences, maxlen=max_length, padding=padding_type, truncating=trunc_type)\n", "print(model.predict(padded))" ], "execution_count": null, "outputs": [] } ] }