{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "accelerator": "GPU", "colab": { "name": "Course 3 - Week 2 - Lesson 3.ipynb", "provenance": [], "collapsed_sections": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.6" } }, "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": "VWI-u75cOh0S" }, "source": [ "\"Open" ] }, { "cell_type": "code", "metadata": { "id": "P-AhVYeBWgQ3" }, "source": [ "# NOTE: PLEASE MAKE SURE YOU ARE RUNNING THIS IN A PYTHON3 ENVIRONMENT\n", "\n", "import tensorflow as tf\n", "print(tf.__version__)\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "wm9S3T8-9H4q" }, "source": [ "# Double check TF 2.0x is installed. If you ran the above block, there was a \n", "# 'reset all runtimes' button at the bottom that you needed to press\n", "import tensorflow as tf\n", "print(tf.__version__)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "_IoM4VFxWpMR" }, "source": [ "# If the import fails, run this\n", "# !pip install -q tensorflow-datasets\n", "\n", "import tensorflow_datasets as tfds\n", "imdb, info = tfds.load(\"imdb_reviews/subwords8k\", with_info=True, as_supervised=True)\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "wHQ2Ko0zl7M4" }, "source": [ "train_data, test_data = imdb['train'], imdb['test']" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "fqGRSe_eCdOz" }, "source": [ "tokenizer = info.features['text'].encoder" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "F6b_EraCDLOh" }, "source": [ "print(tokenizer.subwords)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "fPl2BXhYEHRP" }, "source": [ "sample_string = 'TensorFlow, from basics to mastery'\n", "\n", "tokenized_string = tokenizer.encode(sample_string)\n", "print ('Tokenized string is {}'.format(tokenized_string))\n", "\n", "original_string = tokenizer.decode(tokenized_string)\n", "print ('The original string: {}'.format(original_string))\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "_3t7vvNLEZml" }, "source": [ "for ts in tokenized_string:\n", " print ('{} ----> {}'.format(ts, tokenizer.decode([ts])))" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "j2jS31uHOh0Y" }, "source": [ "BUFFER_SIZE = 10000\n", "BATCH_SIZE = 64\n", "\n", "train_dataset = train_data.shuffle(BUFFER_SIZE)\n", "train_dataset = train_dataset.padded_batch(BATCH_SIZE, tf.compat.v1.data.get_output_shapes(train_dataset))\n", "test_dataset = test_data.padded_batch(BATCH_SIZE, tf.compat.v1.data.get_output_shapes(test_data))" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "5NEpdhb8AxID" }, "source": [ "embedding_dim = 64\n", "model = tf.keras.Sequential([\n", " tf.keras.layers.Embedding(tokenizer.vocab_size, embedding_dim),\n", " tf.keras.layers.GlobalAveragePooling1D(),\n", " tf.keras.layers.Dense(6, activation='relu'),\n", " tf.keras.layers.Dense(1, activation='sigmoid')\n", "])\n", "\n", "model.summary()" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "fkt8c5dNuUlT" }, "source": [ "num_epochs = 10\n", "\n", "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n", "\n", "history = model.fit(train_dataset, epochs=num_epochs, validation_data=test_dataset)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "-_rMnm7WxQGT" }, "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": "qACq5FLzTW4A" }, "source": [ "e = model.layers[0]\n", "weights = e.get_weights()[0]\n", "print(weights.shape) # shape: (vocab_size, embedding_dim)\n", "\n", "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, tokenizer.vocab_size):\n", " word = tokenizer.decode([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()\n", "\n", "\n", "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": [] } ] }