{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "DweYe9FcbMK_" }, "source": [ "##### Copyright 2018 The TensorFlow Authors.\n" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "cellView": "form", "execution": { "iopub.execute_input": "2022-12-15T01:16:30.989655Z", "iopub.status.busy": "2022-12-15T01:16:30.989236Z", "iopub.status.idle": "2022-12-15T01:16:30.993199Z", "shell.execute_reply": "2022-12-15T01:16:30.992581Z" }, "id": "AVV2e0XKbJeX" }, "outputs": [], "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." ] }, { "cell_type": "markdown", "metadata": { "id": "sZfSvVcDo6GQ" }, "source": [ "# テキストを読み込む" ] }, { "cell_type": "markdown", "metadata": { "id": "giK0nMbZFnoR" }, "source": [ "\n", " \n", " \n", " \n", " \n", "
TensorFlow.org で表示 Google Colab で実行 GitHub でソースを表示 ノートブックをダウンロード
" ] }, { "cell_type": "markdown", "metadata": { "id": "dwlfPb11GH8J" }, "source": [ "このチュートリアルでは、テキストを読み込んで前処理する 2 つの方法を紹介します。\n", "\n", "- まず、Keras ユーティリティと前処理レイヤーを使用します。これには、データを `tf.data.Dataset` に変換するための `tf.keras.utils.text_dataset_from_directory` とデータを標準化、トークン化、およびベクトル化するための `tf.keras.layers.TextVectorization` が含まれます。TensorFlow を初めて使用する場合は、これらから始める必要があります。\n", "- 次に、`tf.data.TextLineDataset` などの低レベルのユーティリティを使用してテキストファイルを読み込み、`text.UnicodeScriptTokenizer` や `text.case_fold_utf8` などの [TensorFlow Text](https://www.tensorflow.org/text) APIを使用して、よりきめ細かい制御のためにデータを前処理します。" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:16:30.996662Z", "iopub.status.busy": "2022-12-15T01:16:30.996451Z", "iopub.status.idle": "2022-12-15T01:16:57.693479Z", "shell.execute_reply": "2022-12-15T01:16:57.692557Z" }, "id": "sa6IKWvADqH7" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Collecting tensorflow-text==2.8.*\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Downloading tensorflow_text-2.8.2-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (4.9 MB)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: tensorflow-hub>=0.8.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow-text==2.8.*) (0.12.0)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting tensorflow<2.9,>=2.8.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Downloading tensorflow-2.8.4-cp39-cp39-manylinux2010_x86_64.whl (498.1 MB)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: absl-py>=0.4.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.3.0)\r\n", "Requirement already satisfied: typing-extensions>=3.6.6 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (4.4.0)\r\n", "Requirement already satisfied: flatbuffers>=1.12 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (22.12.6)\r\n", "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.51.1)\r\n", "Requirement already satisfied: numpy>=1.20 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.24.0rc2)\r\n", "Requirement already satisfied: libclang>=9.0.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (14.0.6)\r\n", "Requirement already satisfied: setuptools in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (65.6.3)\r\n", "Requirement already satisfied: google-pasta>=0.1.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.2.0)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting keras<2.9,>=2.8.0rc0\r\n", " Downloading keras-2.8.0-py2.py3-none-any.whl (1.4 MB)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: protobuf<3.20,>=3.9.2 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.19.6)\r\n", "Requirement already satisfied: six>=1.12.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.16.0)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting tensorboard<2.9,>=2.8\r\n", " Downloading tensorboard-2.8.0-py3-none-any.whl (5.8 MB)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: h5py>=2.9.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.7.0)\r\n", "Requirement already satisfied: wrapt>=1.11.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.14.1)\r\n", "Requirement already satisfied: opt-einsum>=2.3.2 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.3.0)\r\n", "Requirement already satisfied: termcolor>=1.1.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.1.1)\r\n", "Requirement already satisfied: astunparse>=1.6.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.6.3)\r\n", "Requirement already satisfied: gast>=0.2.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.4.0)\r\n", "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.28.0)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting keras-preprocessing>=1.1.1\r\n", " Downloading Keras_Preprocessing-1.1.2-py2.py3-none-any.whl (42 kB)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Collecting tensorflow-estimator<2.9,>=2.8\r\n", " Downloading tensorflow_estimator-2.8.0-py2.py3-none-any.whl (462 kB)\r\n", "Requirement already satisfied: wheel<1.0,>=0.23.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from astunparse>=1.6.0->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.38.4)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: google-auth<3,>=1.6.3 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.15.0)\r\n", "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.8.1)\r\n", "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.6.1)\r\n", "Requirement already satisfied: werkzeug>=0.11.15 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.2.2)\r\n", "Requirement already satisfied: requests<3,>=2.21.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.28.1)\r\n", "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.4.6)\r\n", "Requirement already satisfied: markdown>=2.6.8 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.4.1)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: pyasn1-modules>=0.2.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.3.0rc1)\r\n", "Requirement already satisfied: rsa<5,>=3.1.4 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (4.9)\r\n", "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (5.2.0)\r\n", "Requirement already satisfied: requests-oauthlib>=0.7.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.3.1)\r\n", "Requirement already satisfied: importlib-metadata>=4.4 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from markdown>=2.6.8->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (5.1.0)\r\n", "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (1.26.13)\r\n", "Requirement already satisfied: certifi>=2017.4.17 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2022.12.7)\r\n", "Requirement already satisfied: idna<4,>=2.5 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.4)\r\n", "Requirement already satisfied: charset-normalizer<3,>=2 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from requests<3,>=2.21.0->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.1.1)\r\n", "Requirement already satisfied: MarkupSafe>=2.1.1 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from werkzeug>=0.11.15->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (2.1.1)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: zipp>=0.5 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.11.0)\r\n", "Requirement already satisfied: pyasn1<0.6.0,>=0.4.6 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (0.5.0rc2)\r\n", "Requirement already satisfied: oauthlib>=3.0.0 in /tmpfs/src/tf_docs_env/lib/python3.9/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<2.9,>=2.8->tensorflow<2.9,>=2.8.0->tensorflow-text==2.8.*) (3.2.2)\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Installing collected packages: tensorflow-estimator, keras, keras-preprocessing, tensorboard, tensorflow, tensorflow-text\r\n", " Attempting uninstall: tensorflow-estimator\r\n", " Found existing installation: tensorflow-estimator 2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Uninstalling tensorflow-estimator-2.11.0:\r\n", " Successfully uninstalled tensorflow-estimator-2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Attempting uninstall: keras\r\n", " Found existing installation: keras 2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Uninstalling keras-2.11.0:\r\n", " Successfully uninstalled keras-2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Attempting uninstall: tensorboard\r\n", " Found existing installation: tensorboard 2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Uninstalling tensorboard-2.11.0:\r\n", " Successfully uninstalled tensorboard-2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Attempting uninstall: tensorflow\r\n", " Found existing installation: tensorflow 2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Uninstalling tensorflow-2.11.0:\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Successfully uninstalled tensorflow-2.11.0\r\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Successfully installed keras-2.8.0 keras-preprocessing-1.1.2 tensorboard-2.8.0 tensorflow-2.8.4 tensorflow-estimator-2.8.0 tensorflow-text-2.8.2\r\n" ] } ], "source": [ "!pip install \"tensorflow-text==2.8.*\"" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:16:57.698160Z", "iopub.status.busy": "2022-12-15T01:16:57.697592Z", "iopub.status.idle": "2022-12-15T01:17:00.215054Z", "shell.execute_reply": "2022-12-15T01:17:00.214376Z" }, "id": "baYFZMW_bJHh" }, "outputs": [], "source": [ "import collections\n", "import pathlib\n", "\n", "import tensorflow as tf\n", "\n", "from tensorflow.keras import layers\n", "from tensorflow.keras import losses\n", "from tensorflow.keras import utils\n", "from tensorflow.keras.layers import TextVectorization\n", "\n", "import tensorflow_datasets as tfds\n", "import tensorflow_text as tf_text" ] }, { "cell_type": "markdown", "metadata": { "id": "Az-d_K5_HQ5k" }, "source": [ "## 例 1: StackOverflow の質問のタグを予測する\n", "\n", "最初の例として、StackOverflow からプログラミングの質問のデータセットをダウンロードします。それぞれの質問 (「ディクショナリを値で並べ替えるにはどうすればよいですか？」) は、1 つのタグ (Python、`CSharp`、`JavaScript`、または`Java`) でラベルされています。このタスクでは、質問のタグを予測するモデルを開発します。これは、マルチクラス分類の例です。マルチクラス分類は、重要で広く適用できる機械学習の問題です。" ] }, { "cell_type": "markdown", "metadata": { "id": "tjC3yLa5IjP7" }, "source": [ "### データセットをダウンロードして調査する\n", "\n", "まず、`tf.keras.utils.get_file` を使用して Stack Overflow データセットをダウンロードし、ディレクトリの構造を調べます。" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:00.219199Z", "iopub.status.busy": "2022-12-15T01:17:00.218769Z", "iopub.status.idle": "2022-12-15T01:17:02.248984Z", "shell.execute_reply": "2022-12-15T01:17:02.248138Z" }, "id": "8ELgzA6SHTuV" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/stack_overflow_16k.tar.gz\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 16384/6053168 [..............................] - ETA: 0s" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "6053888/6053168 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "6062080/6053168 [==============================] - 0s 0us/step\n" ] } ], "source": [ "data_url = 'https://storage.googleapis.com/download.tensorflow.org/data/stack_overflow_16k.tar.gz'\n", "\n", "dataset_dir = utils.get_file(\n", " origin=data_url,\n", " untar=True,\n", " cache_dir='stack_overflow',\n", " cache_subdir='')\n", "\n", "dataset_dir = pathlib.Path(dataset_dir).parent" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:02.253083Z", "iopub.status.busy": "2022-12-15T01:17:02.252573Z", "iopub.status.idle": "2022-12-15T01:17:02.259634Z", "shell.execute_reply": "2022-12-15T01:17:02.259076Z" }, "id": "jIrPl5fUH2gb" }, "outputs": [ { "data": { "text/plain": [ "[PosixPath('/tmp/.keras/test'),\n", " PosixPath('/tmp/.keras/train'),\n", " PosixPath('/tmp/.keras/README.md'),\n", " PosixPath('/tmp/.keras/stack_overflow_16k.tar.gz')]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "list(dataset_dir.iterdir())" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:02.263050Z", "iopub.status.busy": "2022-12-15T01:17:02.262562Z", "iopub.status.idle": "2022-12-15T01:17:02.266938Z", "shell.execute_reply": "2022-12-15T01:17:02.266329Z" }, "id": "fEoV7YByJoWQ" }, "outputs": [ { "data": { "text/plain": [ "[PosixPath('/tmp/.keras/train/csharp'),\n", " PosixPath('/tmp/.keras/train/javascript'),\n", " PosixPath('/tmp/.keras/train/python'),\n", " PosixPath('/tmp/.keras/train/java')]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "train_dir = dataset_dir/'train'\n", "list(train_dir.iterdir())" ] }, { "cell_type": "markdown", "metadata": { "id": "3mxAN17MhEh0" }, "source": [ "`train/csharp`、`train/java`、`train/python` および `train/javascript` ディレクトリには、多くのテキストファイルが含まれています。それぞれが Stack Overflow の質問です。\n", "\n", "サンプルファイルを出力してデータを調べます。" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:02.270216Z", "iopub.status.busy": "2022-12-15T01:17:02.269781Z", "iopub.status.idle": "2022-12-15T01:17:02.273488Z", "shell.execute_reply": "2022-12-15T01:17:02.272835Z" }, "id": "Go1vTSGdJu08" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "why does this blank program print true x=true.def stupid():. x=false.stupid().print x\n", "\n" ] } ], "source": [ "sample_file = train_dir/'python/1755.txt'\n", "\n", "with open(sample_file) as f:\n", " print(f.read())" ] }, { "cell_type": "markdown", "metadata": { "id": "deWBTkpJiO7D" }, "source": [ "### データセットを読み込む\n", "\n", "次に、データをディスクから読み込み、トレーニングに適した形式に準備します。これを行うには、`tf.keras.utils.text_dataset_from_directory` ユーティリティを使用して、ラベル付きの `tf.data.Dataset` を作成します。これは、入力パイプラインを構築するための強力なツールのコレクションです。`tf.data` を始めて使用する場合は、[tf.data: TensorFlow 入力パイプラインを構築する](../../guide/data.ipynb)を参照してください。\n", "\n", "`tf.keras.utils.text_dataset_from_directory` API は、次のようなディレクトリ構造を想定しています。\n", "\n", "```\n", "train/\n", "...csharp/\n", "......1.txt\n", "......2.txt\n", "...java/\n", "......1.txt\n", "......2.txt\n", "...javascript/\n", "......1.txt\n", "......2.txt\n", "...python/\n", "......1.txt\n", "......2.txt\n", "```" ] }, { "cell_type": "markdown", "metadata": { "id": "Dyl6JTAjlbQV" }, "source": [ "機械学習実験を実行するときは、データセットを[トレーニング](https://developers.google.com/machine-learning/glossary#training_set)、[検証](https://developers.google.com/machine-learning/glossary#validation_set)、および、[テスト](https://developers.google.com/machine-learning/glossary#test-set)の 3 つに分割することをお勧めします。\n", "\n", "Stack Overflow データセットは、すでにトレーニングセットとテストセットに分割されていますが、検証セットはありません。\n", "\n", "`tf.keras.utils.text_dataset_from_directory` を使用し、`validation_split` を `0.2` (20%) に設定し、トレーニングデータを 80:20 に分割して検証セットを作成します。" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:02.276837Z", "iopub.status.busy": "2022-12-15T01:17:02.276334Z", "iopub.status.idle": "2022-12-15T01:17:05.793949Z", "shell.execute_reply": "2022-12-15T01:17:05.793190Z" }, "id": "qqyliMw8N-az" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found 8000 files belonging to 4 classes.\n", "Using 6400 files for training.\n" ] } ], "source": [ "batch_size = 32\n", "seed = 42\n", "\n", "raw_train_ds = utils.text_dataset_from_directory(\n", " train_dir,\n", " batch_size=batch_size,\n", " validation_split=0.2,\n", " subset='training',\n", " seed=seed)" ] }, { "cell_type": "markdown", "metadata": { "id": "DMI_gPLfloD7" }, "source": [ "前のセル出力が示すように、トレーニングフォルダには 8,000 の例があり、そのうち 80％ (6,400) をトレーニングに使用します。`tf.data.Dataset` を `Model.fit` に直接渡すことで、モデルをトレーニングできます。詳細は、後ほど見ていきます。\n", "\n", "まず、データセットを反復処理し、いくつかの例を出力して、データを確認します。\n", "\n", "注意: 分類問題の難易度を上げるために、データセットの作成者は、プログラミングの質問で、*Python*、*CSharp*、*JavaScript*、*Java* という単語を *blank* に置き換えました。" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:05.798150Z", "iopub.status.busy": "2022-12-15T01:17:05.797586Z", "iopub.status.idle": "2022-12-15T01:17:05.840737Z", "shell.execute_reply": "2022-12-15T01:17:05.840052Z" }, "id": "_JMTyZ6Glt_C" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Question: b'\"my tester is going to the wrong constructor i am new to programming so if i ask a question that can be easily fixed, please forgive me. my program has a tester class with a main. when i send that to my regularpolygon class, it sends it to the wrong constructor. i have two constructors. 1 without perameters..public regularpolygon(). {. mynumsides = 5;. mysidelength = 30;. }//end default constructor...and my second, with perameters. ..public regularpolygon(int numsides, double sidelength). {. mynumsides = numsides;. mysidelength = sidelength;. }// end constructor...in my tester class i have these two lines:..regularpolygon shape = new regularpolygon(numsides, sidelength);. shape.menu();...numsides and sidelength were declared and initialized earlier in the testing class...so what i want to happen, is the tester class sends numsides and sidelength to the second constructor and use it in that class. but it only uses the default constructor, which therefor ruins the whole rest of the program. can somebody help me?..for those of you who want to see more of my code: here you go..public double vertexangle(). {. system.out.println(\"\"the vertex angle method: \"\" + mynumsides);// prints out 5. system.out.println(\"\"the vertex angle method: \"\" + mysidelength); // prints out 30.. double vertexangle;. vertexangle = ((mynumsides - 2.0) / mynumsides) * 180.0;. return vertexangle;. }//end method vertexangle..public void menu().{. system.out.println(mynumsides); // prints out what the user puts in. system.out.println(mysidelength); // prints out what the user puts in. gotographic();. calcr(mynumsides, mysidelength);. calcr(mynumsides, mysidelength);. print(); .}// end menu...this is my entire tester class:..public static void main(string[] arg).{. int numsides;. double sidelength;. scanner keyboard = new scanner(system.in);.. system.out.println(\"\"welcome to the regular polygon program!\"\");. system.out.println();.. system.out.print(\"\"enter the number of sides of the polygon ==> \"\");. numsides = keyboard.nextint();. system.out.println();.. system.out.print(\"\"enter the side length of each side ==> \"\");. sidelength = keyboard.nextdouble();. system.out.println();.. regularpolygon shape = new regularpolygon(numsides, sidelength);. shape.menu();.}//end main...for testing it i sent it numsides 4 and sidelength 100.\"\\n'\n", "Label: 1\n", "Question: b'\"blank code slow skin detection this code changes the color space to lab and using a threshold finds the skin area of an image. but it\\'s ridiculously slow. i don\\'t know how to make it faster ? ..from colormath.color_objects import *..def skindetection(img, treshold=80, color=[255,20,147]):.. print img.shape. res=img.copy(). for x in range(img.shape[0]):. for y in range(img.shape[1]):. rgbimg=rgbcolor(img[x,y,0],img[x,y,1],img[x,y,2]). labimg=rgbimg.convert_to(\\'lab\\', debug=false). if (labimg.lab_l > treshold):. res[x,y,:]=color. else: . res[x,y,:]=img[x,y,:].. return res\"\\n'\n", "Label: 3\n", "Question: b'\"option and validation in blank i want to add a new option on my system where i want to add two text files, both rental.txt and customer.txt. inside each text are id numbers of the customer, the videotape they need and the price...i want to place it as an option on my code. right now i have:...add customer.rent return.view list.search.exit...i want to add this as my sixth option. say for example i ordered a video, it would display the price and would let me confirm the price and if i am going to buy it or not...here is my current code:.. import blank.io.*;. import blank.util.arraylist;. import static blank.lang.system.out;.. public class rentalsystem{. static bufferedreader input = new bufferedreader(new inputstreamreader(system.in));. static file file = new file(\"\"file.txt\"\");. static arraylist<string> list = new arraylist<string>();. static int rows;.. public static void main(string[] args) throws exception{. introduction();. system.out.print(\"\"nn\"\");. login();. system.out.print(\"\"nnnnnnnnnnnnnnnnnnnnnn\"\");. introduction();. string repeat;. do{. loadfile();. system.out.print(\"\"nwhat do you want to do?nn\"\");. system.out.print(\"\"n - - - - - - - - - - - - - - - - - - - - - - -\"\");. system.out.print(\"\"nn | 1. add customer | 2. rent return |n\"\");. system.out.print(\"\"n - - - - - - - - - - - - - - - - - - - - - - -\"\");. system.out.print(\"\"nn | 3. view list | 4. search |n\"\");. system.out.print(\"\"n - - - - - - - - - - - - - - - - - - - - - - -\"\");. system.out.print(\"\"nn | 5. exit |n\"\");. system.out.print(\"\"n - - - - - - - - - -\"\");. system.out.print(\"\"nnchoice:\"\");. int choice = integer.parseint(input.readline());. switch(choice){. case 1:. writedata();. break;. case 2:. rentdata();. break;. case 3:. viewlist();. break;. case 4:. search();. break;. case 5:. system.out.println(\"\"goodbye!\"\");. system.exit(0);. default:. system.out.print(\"\"invalid choice: \"\");. break;. }. system.out.print(\"\"ndo another task? [y/n] \"\");. repeat = input.readline();. }while(repeat.equals(\"\"y\"\"));.. if(repeat!=\"\"y\"\") system.out.println(\"\"ngoodbye!\"\");.. }.. public static void writedata() throws exception{. system.out.print(\"\"nname: \"\");. string cname = input.readline();. system.out.print(\"\"address: \"\");. string add = input.readline();. system.out.print(\"\"phone no.: \"\");. string pno = input.readline();. system.out.print(\"\"rental amount: \"\");. string ramount = input.readline();. system.out.print(\"\"tapenumber: \"\");. string tno = input.readline();. system.out.print(\"\"title: \"\");. string title = input.readline();. system.out.print(\"\"date borrowed: \"\");. string dborrowed = input.readline();. system.out.print(\"\"due date: \"\");. string ddate = input.readline();. createline(cname, add, pno, ramount,tno, title, dborrowed, ddate);. rentdata();. }.. public static void createline(string name, string address, string phone , string rental, string tapenumber, string title, string borrowed, string due) throws exception{. filewriter fw = new filewriter(file, true);. fw.write(\"\"nname: \"\"+name + \"\"naddress: \"\" + address +\"\"nphone no.: \"\"+ phone+\"\"nrentalamount: \"\"+rental+\"\"ntape no.: \"\"+ tapenumber+\"\"ntitle: \"\"+ title+\"\"ndate borrowed: \"\"+borrowed +\"\"ndue date: \"\"+ due+\"\":rn\"\");. fw.close();. }.. public static void loadfile() throws exception{. try{. list.clear();. fileinputstream fstream = new fileinputstream(file);. bufferedreader br = new bufferedreader(new inputstreamreader(fstream));. rows = 0;. while( br.ready()). {. list.add(br.readline());. rows++;. }. br.close();. } catch(exception e){. system.out.println(\"\"list not yet loaded.\"\");. }. }.. public static void viewlist(){. system.out.print(\"\"n~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. system.out.print(\"\" |list of all costumers|\"\");. system.out.print(\"\"~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. for(int i = 0; i <rows; i++){. system.out.println(list.get(i));. }. }. public static void rentdata()throws exception. { system.out.print(\"\"n~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. system.out.print(\"\" |rent data list|\"\");. system.out.print(\"\"~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. system.out.print(\"\"nenter customer name: \"\");. string cname = input.readline();. system.out.print(\"\"date borrowed: \"\");. string dborrowed = input.readline();. system.out.print(\"\"due date: \"\");. string ddate = input.readline();. system.out.print(\"\"return date: \"\");. string rdate = input.readline();. system.out.print(\"\"rent amount: \"\");. string ramount = input.readline();.. system.out.print(\"\"you pay:\"\"+ramount);... }. public static void search()throws exception. { system.out.print(\"\"n~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. system.out.print(\"\" |search costumers|\"\");. system.out.print(\"\"~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~\"\");. system.out.print(\"\"nenter costumer name: \"\");. string cname = input.readline();. boolean found = false;.. for(int i=0; i < rows; i++){. string temp[] = list.get(i).split(\"\",\"\");.. if(cname.equals(temp[0])){. system.out.println(\"\"search result:nyou are \"\" + temp[0] + \"\" from \"\" + temp[1] + \"\".\"\"+ temp[2] + \"\".\"\"+ temp[3] + \"\".\"\"+ temp[4] + \"\".\"\"+ temp[5] + \"\" is \"\" + temp[6] + \"\".\"\"+ temp[7] + \"\" is \"\" + temp[8] + \"\".\"\");. found = true;. }. }.. if(!found){. system.out.print(\"\"no results.\"\");. }.. }.. public static boolean evaluate(string uname, string pass){. if (uname.equals(\"\"admin\"\")&&pass.equals(\"\"12345\"\")) return true;. else return false;. }.. public static string login()throws exception{. bufferedreader input=new bufferedreader(new inputstreamreader(system.in));. int counter=0;. do{. system.out.print(\"\"username:\"\");. string uname =input.readline();. system.out.print(\"\"password:\"\");. string pass =input.readline();.. boolean accept= evaluate(uname,pass);.. if(accept){. break;. }else{. system.out.println(\"\"incorrect username or password!\"\");. counter ++;. }. }while(counter<3);.. if(counter !=3) return \"\"login successful\"\";. else return \"\"login failed\"\";. }. public static void introduction() throws exception{.. system.out.println(\"\" - - - - - - - - - - - - - - - - - - - - - - - - -\"\");. system.out.println(\"\" ! r e n t a l !\"\");. system.out.println(\"\" ! ~ ~ ~ ~ ~ ! ================= ! ~ ~ ~ ~ ~ !\"\");. system.out.println(\"\" ! s y s t e m !\"\");. system.out.println(\"\" - - - - - - - - - - - - - - - - - - - - - - - - -\"\");. }..}\"\\n'\n", "Label: 1\n", "Question: b'\"exception: dynamic sql generation for the updatecommand is not supported against a selectcommand that does not return any key i dont know what is the problem this my code : ..string nomtable;..datatable listeetablissementtable = new datatable();.datatable listeinteretstable = new datatable();.dataset ds = new dataset();.sqldataadapter da;.sqlcommandbuilder cmdb;..private void listeinterets_click(object sender, eventargs e).{. nomtable = \"\"listeinteretstable\"\";. d.cnx.open();. da = new sqldataadapter(\"\"select nome from offices\"\", d.cnx);. ds = new dataset();. da.fill(ds, nomtable);. datagridview1.datasource = ds.tables[nomtable];.}..private void sauvgarder_click(object sender, eventargs e).{. d.cnx.open();. cmdb = new sqlcommandbuilder(da);. da.update(ds, nomtable);. d.cnx.close();.}\"\\n'\n", "Label: 0\n", "Question: b'\"parameter with question mark and super in blank, i\\'ve come across a method that is formatted like this:..public final subscription subscribe(final action1<? super t> onnext, final action1<throwable> onerror) {.}...in the first parameter, what does the question mark and super mean?\"\\n'\n", "Label: 1\n", "Question: b'call two objects wsdl the first time i got a very strange wsdl. ..i would like to call the object (interface - invoicecheck_out) do you know how?....i would like to call the object (variable) do you know how?..try to call (it`s ok)....try to call (how call this?)\\n'\n", "Label: 0\n", "Question: b\"how to correctly make the icon for systemtray in blank using icon sizes of any dimension for systemtray doesn't look good overall. .what is the correct way of making icons for windows system tray?..screenshots: http://imgur.com/zsibwn9..icon: http://imgur.com/vsh4zo8\\n\"\n", "Label: 0\n", "Question: b'\"is there a way to check a variable that exists in a different script than the original one? i\\'m trying to check if a variable, which was previously set to true in 2.py in 1.py, as 1.py is only supposed to continue if the variable is true...2.py..import os..completed = false..#some stuff here..completed = true...1.py..import 2 ..if completed == true. #do things...however i get a syntax error at ..if completed == true\"\\n'\n", "Label: 3\n", "Question: b'\"blank control flow i made a number which asks for 2 numbers with blank and responds with the corresponding message for the case. how come it doesnt work for the second number ? .regardless what i enter for the second number , i am getting the message \"\"your number is in the range 0-10\"\"...using system;.using system.collections.generic;.using system.linq;.using system.text;..namespace consoleapplication1.{. class program. {. static void main(string[] args). {. string myinput; // declaring the type of the variables. int myint;.. string number1;. int number;... console.writeline(\"\"enter a number\"\");. myinput = console.readline(); //muyinput is a string which is entry input. myint = int32.parse(myinput); // myint converts the string into an integer.. if (myint > 0). console.writeline(\"\"your number {0} is greater than zero.\"\", myint);. else if (myint < 0). console.writeline(\"\"your number {0} is less than zero.\"\", myint);. else. console.writeline(\"\"your number {0} is equal zero.\"\", myint);.. console.writeline(\"\"enter another number\"\");. number1 = console.readline(); . number = int32.parse(myinput); .. if (number < 0 || number == 0). console.writeline(\"\"your number {0} is less than zero or equal zero.\"\", number);. else if (number > 0 && number <= 10). console.writeline(\"\"your number {0} is in the range from 0 to 10.\"\", number);. else. console.writeline(\"\"your number {0} is greater than 10.\"\", number);.. console.writeline(\"\"enter another number\"\");.. }. } .}\"\\n'\n", "Label: 0\n", "Question: b'\"credentials cannot be used for ntlm authentication i am getting org.apache.commons.httpclient.auth.invalidcredentialsexception: credentials cannot be used for ntlm authentication: exception in eclipse..whether it is possible mention eclipse to take system proxy settings directly?..public class httpgetproxy {. private static final string proxy_host = \"\"proxy.****.com\"\";. private static final int proxy_port = 6050;.. public static void main(string[] args) {. httpclient client = new httpclient();. httpmethod method = new getmethod(\"\"https://kodeblank.org\"\");.. hostconfiguration config = client.gethostconfiguration();. config.setproxy(proxy_host, proxy_port);.. string username = \"\"*****\"\";. string password = \"\"*****\"\";. credentials credentials = new usernamepasswordcredentials(username, password);. authscope authscope = new authscope(proxy_host, proxy_port);.. client.getstate().setproxycredentials(authscope, credentials);.. try {. client.executemethod(method);.. if (method.getstatuscode() == httpstatus.sc_ok) {. string response = method.getresponsebodyasstring();. system.out.println(\"\"response = \"\" + response);. }. } catch (ioexception e) {. e.printstacktrace();. } finally {. method.releaseconnection();. }. }.}...exception:... dec 08, 2017 1:41:39 pm . org.apache.commons.httpclient.auth.authchallengeprocessor selectauthscheme. info: ntlm authentication scheme selected. dec 08, 2017 1:41:39 pm org.apache.commons.httpclient.httpmethoddirector executeconnect. severe: credentials cannot be used for ntlm authentication: . org.apache.commons.httpclient.usernamepasswordcredentials. org.apache.commons.httpclient.auth.invalidcredentialsexception: credentials . cannot be used for ntlm authentication: . enter code here . org.apache.commons.httpclient.usernamepasswordcredentials. at org.apache.commons.httpclient.auth.ntlmscheme.authenticate(ntlmscheme.blank:332). at org.apache.commons.httpclient.httpmethoddirector.authenticateproxy(httpmethoddirector.blank:320). at org.apache.commons.httpclient.httpmethoddirector.executeconnect(httpmethoddirector.blank:491). at org.apache.commons.httpclient.httpmethoddirector.executewithretry(httpmethoddirector.blank:391). at org.apache.commons.httpclient.httpmethoddirector.executemethod(httpmethoddirector.blank:171). at org.apache.commons.httpclient.httpclient.executemethod(httpclient.blank:397). at org.apache.commons.httpclient.httpclient.executemethod(httpclient.blank:323). at httpgetproxy.main(httpgetproxy.blank:31). dec 08, 2017 1:41:39 pm org.apache.commons.httpclient.httpmethoddirector processproxyauthchallenge. info: failure authenticating with ntlm @proxy.****.com:6050\"\\n'\n", "Label: 1\n" ] } ], "source": [ "for text_batch, label_batch in raw_train_ds.take(1):\n", " for i in range(10):\n", " print(\"Question: \", text_batch.numpy()[i])\n", " print(\"Label:\", label_batch.numpy()[i])" ] }, { "cell_type": "markdown", "metadata": { "id": "jCZGl4Q5l2sS" }, "source": [ "ラベルは、`0`、`1`、`2` または `3` です。これらのどれがどの文字列ラベルに対応するかを確認するには、データセットの `class_names` プロパティを確認します。\n" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:05.843906Z", "iopub.status.busy": "2022-12-15T01:17:05.843631Z", "iopub.status.idle": "2022-12-15T01:17:05.847545Z", "shell.execute_reply": "2022-12-15T01:17:05.846896Z" }, "id": "gIpCS7YjmGkj" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Label 0 corresponds to csharp\n", "Label 1 corresponds to java\n", "Label 2 corresponds to javascript\n", "Label 3 corresponds to python\n" ] } ], "source": [ "for i, label in enumerate(raw_train_ds.class_names):\n", " print(\"Label\", i, \"corresponds to\", label)" ] }, { "cell_type": "markdown", "metadata": { "id": "oUsdn-37qol9" }, "source": [ "次に、`tf.keras.utils.text_dataset_from_directory` を使って検証およびテスト用データセットを作成します。トレーニング用セットの残りの 1,600 件のレビューを検証に使用します。\n", "\n", "注意: `tf.keras.utils.text_dataset_from_directory` の `validation_split` および `subset` 引数を使用する場合は、必ずランダムシードを指定するか、`shuffle=False`を渡して、検証とトレーニング分割に重複がないようにします。" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:05.850951Z", "iopub.status.busy": "2022-12-15T01:17:05.850399Z", "iopub.status.idle": "2022-12-15T01:17:06.037601Z", "shell.execute_reply": "2022-12-15T01:17:06.036844Z" }, "id": "x7m6sCWJQuYt" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found 8000 files belonging to 4 classes.\n", "Using 1600 files for validation.\n" ] } ], "source": [ "# Create a validation set.\n", "raw_val_ds = utils.text_dataset_from_directory(\n", " train_dir,\n", " batch_size=batch_size,\n", " validation_split=0.2,\n", " subset='validation',\n", " seed=seed)" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:06.041570Z", "iopub.status.busy": "2022-12-15T01:17:06.040891Z", "iopub.status.idle": "2022-12-15T01:17:06.232740Z", "shell.execute_reply": "2022-12-15T01:17:06.232110Z" }, "id": "BXMZc7fMQwKE" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found 8000 files belonging to 4 classes.\n" ] } ], "source": [ "test_dir = dataset_dir/'test'\n", "\n", "# Create a test set.\n", "raw_test_ds = utils.text_dataset_from_directory(\n", " test_dir,\n", " batch_size=batch_size)" ] }, { "cell_type": "markdown", "metadata": { "id": "Xdt-ATrGRGDL" }, "source": [ "### トレーニング用データセットを準備する" ] }, { "cell_type": "markdown", "metadata": { "id": "N6fRti45Rlj8" }, "source": [ "次に、`tf.keras.layers.TextVectorization` レイヤーを使用して、データを標準化、トークン化、およびベクトル化します。\n", "\n", "- 標準化とは、テキストを前処理することを指します。通常、句読点や HTML 要素を削除して、データセットを簡素化します。\n", "- トークン化とは、文字列をトークンに分割することです（たとえば、空白で分割することにより、文を個々の単語に分割します）。\n", "- ベクトル化とは、トークンを数値に変換して、ニューラルネットワークに入力できるようにすることです。\n", "\n", "これらのタスクはすべて、このレイヤーで実行できます。これらの詳細については、`tf.keras.layers.TextVectorization` API ドキュメントを参照してください。\n", "\n", "注意点 :\n", "\n", "- デフォルトの標準化では、テキストが小文字に変換され、句読点が削除されます (`standardize='lower_and_strip_punctuation'`)。\n", "- デフォルトのトークナイザーは空白で分割されます (`split='whitespace'`)。\n", "- デフォルトのベクトル化モードは `int` です (`output_mode='int'`)。これは整数インデックスを出力します（トークンごとに1つ）。このモードは、語順を考慮したモデルを構築するために使用できます。`binary` などの他のモードを使用して、[bag-of-word](https://developers.google.com/machine-learning/glossary#bag-of-words) モデルを構築することもできます。\n", "\n", "`TextVectorization` を使用した標準化、トークン化、およびベクトル化について詳しくみるために、2 つのモデルを作成します。\n", "\n", "- まず、`'binary'` ベクトル化モードを使用して、bag-of-words モデルを構築します。\n", "- 次に、1D ConvNet で `'int'` モードを使用します。" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:06.237030Z", "iopub.status.busy": "2022-12-15T01:17:06.236313Z", "iopub.status.idle": "2022-12-15T01:17:06.247659Z", "shell.execute_reply": "2022-12-15T01:17:06.247016Z" }, "id": "voaC43rZR0jc" }, "outputs": [], "source": [ "VOCAB_SIZE = 10000\n", "\n", "binary_vectorize_layer = TextVectorization(\n", " max_tokens=VOCAB_SIZE,\n", " output_mode='binary')" ] }, { "cell_type": "markdown", "metadata": { "id": "ifDPFxuf2Hfz" }, "source": [ "`'int'` モードの場合、最大語彙サイズに加えて、明示的な最大シーケンス長 (`MAX_SEQUENCE_LENGTH`) を設定する必要があります。これにより、レイヤーはシーケンスを正確に `output_sequence_length` 値にパディングまたは切り捨てます。" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:06.251335Z", "iopub.status.busy": "2022-12-15T01:17:06.250870Z", "iopub.status.idle": "2022-12-15T01:17:06.256512Z", "shell.execute_reply": "2022-12-15T01:17:06.255904Z" }, "id": "XWsY01Zl2aRe" }, "outputs": [], "source": [ "MAX_SEQUENCE_LENGTH = 250\n", "\n", "int_vectorize_layer = TextVectorization(\n", " max_tokens=VOCAB_SIZE,\n", " output_mode='int',\n", " output_sequence_length=MAX_SEQUENCE_LENGTH)" ] }, { "cell_type": "markdown", "metadata": { "id": "ts6h9b5atD-Y" }, "source": [ "次に、`TextVectorization.adapt` を呼び出して、前処理レイヤーの状態をデータセットに適合させます。これにより、モデルは文字列から整数へのインデックスを作成します。\n", "\n", "注意: `TextVectorization.adapt` を呼び出すときは、トレーニング用データのみを使用することが重要です (テスト用セットを使用すると情報が漏洩します)。" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:06.259904Z", "iopub.status.busy": "2022-12-15T01:17:06.259410Z", "iopub.status.idle": "2022-12-15T01:17:07.934491Z", "shell.execute_reply": "2022-12-15T01:17:07.933587Z" }, "id": "yTXsdDEqSf9e" }, "outputs": [], "source": [ "# Make a text-only dataset (without labels), then call `TextVectorization.adapt`.\n", "train_text = raw_train_ds.map(lambda text, labels: text)\n", "binary_vectorize_layer.adapt(train_text)\n", "int_vectorize_layer.adapt(train_text)" ] }, { "cell_type": "markdown", "metadata": { "id": "XKVO6Jg7Sls0" }, "source": [ "これらのレイヤーを使用してデータを前処理した結果を出力します。" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:07.939165Z", "iopub.status.busy": "2022-12-15T01:17:07.938549Z", "iopub.status.idle": "2022-12-15T01:17:07.942562Z", "shell.execute_reply": "2022-12-15T01:17:07.941908Z" }, "id": "RngfPyArSsvM" }, "outputs": [], "source": [ "def binary_vectorize_text(text, label):\n", " text = tf.expand_dims(text, -1)\n", " return binary_vectorize_layer(text), label" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:07.946144Z", "iopub.status.busy": "2022-12-15T01:17:07.945642Z", "iopub.status.idle": "2022-12-15T01:17:07.949532Z", "shell.execute_reply": "2022-12-15T01:17:07.948917Z" }, "id": "_1W54wf0LhQ0" }, "outputs": [], "source": [ "def int_vectorize_text(text, label):\n", " text = tf.expand_dims(text, -1)\n", " return int_vectorize_layer(text), label" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:07.952521Z", "iopub.status.busy": "2022-12-15T01:17:07.952265Z", "iopub.status.idle": "2022-12-15T01:17:07.973640Z", "shell.execute_reply": "2022-12-15T01:17:07.972911Z" }, "id": "Vi_sElMiSmXe" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Question tf.Tensor(b'\"what is the difference between these two ways to create an element? var a = document.createelement(\\'div\\');..a.id = \"\"mydiv\"\";...and..var a = document.createelement(\\'div\\').id = \"\"mydiv\"\";...what is the difference between them such that the first one works and the second one doesn\\'t?\"\\n', shape=(), dtype=string)\n", "Label tf.Tensor(2, shape=(), dtype=int32)\n" ] } ], "source": [ "# Retrieve a batch (of 32 reviews and labels) from the dataset.\n", "text_batch, label_batch = next(iter(raw_train_ds))\n", "first_question, first_label = text_batch[0], label_batch[0]\n", "print(\"Question\", first_question)\n", "print(\"Label\", first_label)" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:07.977081Z", "iopub.status.busy": "2022-12-15T01:17:07.976412Z", "iopub.status.idle": "2022-12-15T01:17:07.991716Z", "shell.execute_reply": "2022-12-15T01:17:07.991044Z" }, "id": "UGukZoYv2v3v" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "'binary' vectorized question: tf.Tensor([[1. 1. 0. ... 0. 0. 0.]], shape=(1, 10000), dtype=float32)\n" ] } ], "source": [ "print(\"'binary' vectorized question:\",\n", " binary_vectorize_text(first_question, first_label)[0])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:07.995358Z", "iopub.status.busy": "2022-12-15T01:17:07.994628Z", "iopub.status.idle": "2022-12-15T01:17:08.004456Z", "shell.execute_reply": "2022-12-15T01:17:08.003781Z" }, "id": "Lu07FsIw2yH5" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "'int' vectorized question: tf.Tensor(\n", "[[ 55 6 2 410 211 229 121 895 4 124 32 245 43 5 1 1 5 1\n", " 1 6 2 410 211 191 318 14 2 98 71 188 8 2 199 71 178 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n", " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]], shape=(1, 250), dtype=int64)\n" ] } ], "source": [ "print(\"'int' vectorized question:\",\n", " int_vectorize_text(first_question, first_label)[0])" ] }, { "cell_type": "markdown", "metadata": { "id": "wgjeF9PdS7tN" }, "source": [ "上に示したように、`TextVectorization` の `'binary'` モードは、入力に少なくとも 1 回存在するトークンを示す配列を返しますが、`'int'` モードでは、各トークンが整数に置き換えられるため、トークンの順序が保持されます。\n", "\n", "レイヤーで `TextVectorization.get_vocabulary` を呼び出すことにより、各整数が対応するトークン (文字列) を検索できます。" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:08.007842Z", "iopub.status.busy": "2022-12-15T01:17:08.007328Z", "iopub.status.idle": "2022-12-15T01:17:08.051230Z", "shell.execute_reply": "2022-12-15T01:17:08.050601Z" }, "id": "WpBnTZilS8wt" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1289 ---> roman\n", "313 ---> source\n", "Vocabulary size: 10000\n" ] } ], "source": [ "print(\"1289 ---> \", int_vectorize_layer.get_vocabulary()[1289])\n", "print(\"313 ---> \", int_vectorize_layer.get_vocabulary()[313])\n", "print(\"Vocabulary size: {}\".format(len(int_vectorize_layer.get_vocabulary())))" ] }, { "cell_type": "markdown", "metadata": { "id": "0kHgPE_YwHvp" }, "source": [ "モデルをトレーニングする準備がほぼ整いました。\n", "\n", "最後の前処理ステップとして、トレーニング、検証、およびデータセットのテストのために前に作成した `TextVectorization` レイヤーを適用します。" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:08.054585Z", "iopub.status.busy": "2022-12-15T01:17:08.054128Z", "iopub.status.idle": "2022-12-15T01:17:08.286547Z", "shell.execute_reply": "2022-12-15T01:17:08.285851Z" }, "id": "46LeHmnD55wJ" }, "outputs": [], "source": [ "binary_train_ds = raw_train_ds.map(binary_vectorize_text)\n", "binary_val_ds = raw_val_ds.map(binary_vectorize_text)\n", "binary_test_ds = raw_test_ds.map(binary_vectorize_text)\n", "\n", "int_train_ds = raw_train_ds.map(int_vectorize_text)\n", "int_val_ds = raw_val_ds.map(int_vectorize_text)\n", "int_test_ds = raw_test_ds.map(int_vectorize_text)" ] }, { "cell_type": "markdown", "metadata": { "id": "NHuAF8hYfP5Z" }, "source": [ "### パフォーマンスのためにデータセットを構成する\n", "\n", "以下は、データを読み込むときに I/O がブロックされないようにするために使用する必要がある 2 つの重要な方法です。\n", "\n", "- `Dataset.cache` はデータをディスクから読み込んだ後、データをメモリに保持します。これにより、モデルのトレーニング中にデータセットがボトルネックになることを回避できます。データセットが大きすぎてメモリに収まらない場合は、この方法を使用して、パフォーマンスの高いオンディスクキャッシュを作成することもできます。これは、多くの小さなファイルを読み込むより効率的です。\n", "- `Dataset.prefetch` はトレーニング中にデータの前処理とモデルの実行をオーバーラップさせます。\n", "\n", "以上の 2 つの方法とデータをディスクにキャッシュする方法についての詳細は、データパフォーマンスガイドの プリフェッチを参照してください。" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:08.290860Z", "iopub.status.busy": "2022-12-15T01:17:08.290321Z", "iopub.status.idle": "2022-12-15T01:17:08.293841Z", "shell.execute_reply": "2022-12-15T01:17:08.293285Z" }, "id": "PabA9DFIfSz7" }, "outputs": [], "source": [ "AUTOTUNE = tf.data.AUTOTUNE\n", "\n", "def configure_dataset(dataset):\n", " return dataset.cache().prefetch(buffer_size=AUTOTUNE)" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:08.296909Z", "iopub.status.busy": "2022-12-15T01:17:08.296427Z", "iopub.status.idle": "2022-12-15T01:17:08.303596Z", "shell.execute_reply": "2022-12-15T01:17:08.303006Z" }, "id": "J8GcJLvb3JH0" }, "outputs": [], "source": [ "binary_train_ds = configure_dataset(binary_train_ds)\n", "binary_val_ds = configure_dataset(binary_val_ds)\n", "binary_test_ds = configure_dataset(binary_test_ds)\n", "\n", "int_train_ds = configure_dataset(int_train_ds)\n", "int_val_ds = configure_dataset(int_val_ds)\n", "int_test_ds = configure_dataset(int_test_ds)" ] }, { "cell_type": "markdown", "metadata": { "id": "NYGb7z_bfpGm" }, "source": [ "### モデルをトレーニングする\n", "\n", "ニューラルネットワークを作成します。\n", "\n", "`'binary'` のベクトル化されたデータの場合、単純な bag-of-words 線形モデルを定義し、それを構成してトレーニングします。" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:08.307086Z", "iopub.status.busy": "2022-12-15T01:17:08.306588Z", "iopub.status.idle": "2022-12-15T01:17:15.127703Z", "shell.execute_reply": "2022-12-15T01:17:15.126905Z" }, "id": "2q8iAU-VMzaN" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 2:25 - loss: 1.3763 - accuracy: 0.2188" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 18/200 [=>............................] - ETA: 0s - loss: 1.3713 - accuracy: 0.2743 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 36/200 [====>.........................] - ETA: 0s - loss: 1.3472 - accuracy: 0.3628" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 54/200 [=======>......................] - ETA: 0s - loss: 1.3138 - accuracy: 0.4427" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 71/200 [=========>....................] - ETA: 0s - loss: 1.2850 - accuracy: 0.4934" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 89/200 [============>.................] - ETA: 0s - loss: 1.2614 - accuracy: 0.5302" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "107/200 [===============>..............] - ETA: 0s - loss: 1.2328 - accuracy: 0.5669" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "125/200 [=================>............] - ETA: 0s - loss: 1.2099 - accuracy: 0.5860" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "143/200 [====================>.........] - ETA: 0s - loss: 1.1889 - accuracy: 0.6045" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "160/200 [=======================>......] - ETA: 0s - loss: 1.1644 - accuracy: 0.6203" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "177/200 [=========================>....] - ETA: 0s - loss: 1.1443 - accuracy: 0.6308" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "195/200 [============================>.] - ETA: 0s - loss: 1.1273 - accuracy: 0.6409" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 2s 4ms/step - loss: 1.1221 - accuracy: 0.6439 - val_loss: 0.9157 - val_accuracy: 0.7738\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 2/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.8837 - accuracy: 0.7500" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 26/200 [==>...........................] - ETA: 0s - loss: 0.8588 - accuracy: 0.8293" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 51/200 [======>.......................] - ETA: 0s - loss: 0.8524 - accuracy: 0.8199" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 76/200 [==========>...................] - ETA: 0s - loss: 0.8350 - accuracy: 0.8215" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "100/200 [==============>...............] - ETA: 0s - loss: 0.8240 - accuracy: 0.8200" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "123/200 [=================>............] - ETA: 0s - loss: 0.8155 - accuracy: 0.8176" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "145/200 [====================>.........] - ETA: 0s - loss: 0.8067 - accuracy: 0.8192" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "169/200 [========================>.....] - ETA: 0s - loss: 0.7920 - accuracy: 0.8203" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "192/200 [===========================>..] - ETA: 0s - loss: 0.7840 - accuracy: 0.8210" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.7801 - accuracy: 0.8213 - val_loss: 0.7511 - val_accuracy: 0.7956\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 3/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.7547 - accuracy: 0.7812" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 24/200 [==>...........................] - ETA: 0s - loss: 0.6810 - accuracy: 0.8529" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 47/200 [======>.......................] - ETA: 0s - loss: 0.6778 - accuracy: 0.8531" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 70/200 [=========>....................] - ETA: 0s - loss: 0.6601 - accuracy: 0.8580" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 93/200 [============>.................] - ETA: 0s - loss: 0.6593 - accuracy: 0.8528" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "117/200 [================>.............] - ETA: 0s - loss: 0.6491 - accuracy: 0.8552" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "141/200 [====================>.........] - ETA: 0s - loss: 0.6458 - accuracy: 0.8566" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "166/200 [=======================>......] - ETA: 0s - loss: 0.6372 - accuracy: 0.8575" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "189/200 [===========================>..] - ETA: 0s - loss: 0.6319 - accuracy: 0.8586" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.6284 - accuracy: 0.8589 - val_loss: 0.6653 - val_accuracy: 0.8069\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 4/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.6778 - accuracy: 0.8438" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 24/200 [==>...........................] - ETA: 0s - loss: 0.5755 - accuracy: 0.8841" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 48/200 [======>.......................] - ETA: 0s - loss: 0.5726 - accuracy: 0.8828" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 72/200 [=========>....................] - ETA: 0s - loss: 0.5577 - accuracy: 0.8837" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 96/200 [=============>................] - ETA: 0s - loss: 0.5559 - accuracy: 0.8809" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "120/200 [=================>............] - ETA: 0s - loss: 0.5493 - accuracy: 0.8820" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "144/200 [====================>.........] - ETA: 0s - loss: 0.5474 - accuracy: 0.8837" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "169/200 [========================>.....] - ETA: 0s - loss: 0.5401 - accuracy: 0.8854" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "194/200 [============================>.] - ETA: 0s - loss: 0.5366 - accuracy: 0.8863" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.5349 - accuracy: 0.8863 - val_loss: 0.6118 - val_accuracy: 0.8200\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 5/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.6206 - accuracy: 0.8438" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 25/200 [==>...........................] - ETA: 0s - loss: 0.5015 - accuracy: 0.9062" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 49/200 [======>.......................] - ETA: 0s - loss: 0.4981 - accuracy: 0.9069" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 74/200 [==========>...................] - ETA: 0s - loss: 0.4861 - accuracy: 0.9046" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 99/200 [=============>................] - ETA: 0s - loss: 0.4840 - accuracy: 0.9009" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "124/200 [=================>............] - ETA: 0s - loss: 0.4807 - accuracy: 0.8994" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "146/200 [====================>.........] - ETA: 0s - loss: 0.4789 - accuracy: 0.9011" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "169/200 [========================>.....] - ETA: 0s - loss: 0.4731 - accuracy: 0.9027" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "193/200 [===========================>..] - ETA: 0s - loss: 0.4706 - accuracy: 0.9043" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.4688 - accuracy: 0.9045 - val_loss: 0.5751 - val_accuracy: 0.8281\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 6/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.5743 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 24/200 [==>...........................] - ETA: 0s - loss: 0.4491 - accuracy: 0.9141" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 47/200 [======>.......................] - ETA: 0s - loss: 0.4461 - accuracy: 0.9195" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 70/200 [=========>....................] - ETA: 0s - loss: 0.4319 - accuracy: 0.9192" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 93/200 [============>.................] - ETA: 0s - loss: 0.4340 - accuracy: 0.9143" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "116/200 [================>.............] - ETA: 0s - loss: 0.4267 - accuracy: 0.9138" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "140/200 [====================>.........] - ETA: 0s - loss: 0.4268 - accuracy: 0.9136" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "163/200 [=======================>......] - ETA: 0s - loss: 0.4225 - accuracy: 0.9147" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "187/200 [===========================>..] - ETA: 0s - loss: 0.4195 - accuracy: 0.9151" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.4184 - accuracy: 0.9153 - val_loss: 0.5485 - val_accuracy: 0.8331\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 7/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.5349 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 25/200 [==>...........................] - ETA: 0s - loss: 0.4048 - accuracy: 0.9275" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 49/200 [======>.......................] - ETA: 0s - loss: 0.4007 - accuracy: 0.9305" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 73/200 [=========>....................] - ETA: 0s - loss: 0.3906 - accuracy: 0.9294" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 96/200 [=============>................] - ETA: 0s - loss: 0.3908 - accuracy: 0.9242" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "120/200 [=================>............] - ETA: 0s - loss: 0.3860 - accuracy: 0.9245" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "144/200 [====================>.........] - ETA: 0s - loss: 0.3853 - accuracy: 0.9249" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "168/200 [========================>.....] - ETA: 0s - loss: 0.3821 - accuracy: 0.9252" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "192/200 [===========================>..] - ETA: 0s - loss: 0.3798 - accuracy: 0.9277" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.3782 - accuracy: 0.9281 - val_loss: 0.5284 - val_accuracy: 0.8363\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 8/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.5008 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 23/200 [==>...........................] - ETA: 0s - loss: 0.3678 - accuracy: 0.9321" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 45/200 [=====>........................] - ETA: 0s - loss: 0.3641 - accuracy: 0.9361" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 67/200 [=========>....................] - ETA: 0s - loss: 0.3549 - accuracy: 0.9352" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 89/200 [============>.................] - ETA: 0s - loss: 0.3595 - accuracy: 0.9298" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "111/200 [===============>..............] - ETA: 0s - loss: 0.3523 - accuracy: 0.9319" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "132/200 [==================>...........] - ETA: 0s - loss: 0.3504 - accuracy: 0.9318" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "153/200 [=====================>........] - ETA: 0s - loss: 0.3494 - accuracy: 0.9322" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "177/200 [=========================>....] - ETA: 0s - loss: 0.3447 - accuracy: 0.9345" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.3449 - accuracy: 0.9350 - val_loss: 0.5128 - val_accuracy: 0.8381\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 9/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.4705 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 26/200 [==>...........................] - ETA: 0s - loss: 0.3367 - accuracy: 0.9387" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 50/200 [======>.......................] - ETA: 0s - loss: 0.3334 - accuracy: 0.9406" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 74/200 [==========>...................] - ETA: 0s - loss: 0.3267 - accuracy: 0.9405" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 99/200 [=============>................] - ETA: 0s - loss: 0.3256 - accuracy: 0.9388" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "123/200 [=================>............] - ETA: 0s - loss: 0.3245 - accuracy: 0.9380" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "147/200 [=====================>........] - ETA: 0s - loss: 0.3223 - accuracy: 0.9394" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "171/200 [========================>.....] - ETA: 0s - loss: 0.3190 - accuracy: 0.9412" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "195/200 [============================>.] - ETA: 0s - loss: 0.3177 - accuracy: 0.9425" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.3166 - accuracy: 0.9425 - val_loss: 0.5005 - val_accuracy: 0.8425\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 10/10\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.4433 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 25/200 [==>...........................] - ETA: 0s - loss: 0.3147 - accuracy: 0.9400" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 49/200 [======>.......................] - ETA: 0s - loss: 0.3098 - accuracy: 0.9458" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 73/200 [=========>....................] - ETA: 0s - loss: 0.3016 - accuracy: 0.9473" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 98/200 [=============>................] - ETA: 0s - loss: 0.3009 - accuracy: 0.9464" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "122/200 [=================>............] - ETA: 0s - loss: 0.2994 - accuracy: 0.9454" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "146/200 [====================>.........] - ETA: 0s - loss: 0.2976 - accuracy: 0.9461" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "170/200 [========================>.....] - ETA: 0s - loss: 0.2948 - accuracy: 0.9472" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "194/200 [============================>.] - ETA: 0s - loss: 0.2936 - accuracy: 0.9485" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 3ms/step - loss: 0.2923 - accuracy: 0.9486 - val_loss: 0.4908 - val_accuracy: 0.8413\n" ] } ], "source": [ "binary_model = tf.keras.Sequential([layers.Dense(4)])\n", "\n", "binary_model.compile(\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=True),\n", " optimizer='adam',\n", " metrics=['accuracy'])\n", "\n", "history = binary_model.fit(\n", " binary_train_ds, validation_data=binary_val_ds, epochs=10)" ] }, { "cell_type": "markdown", "metadata": { "id": "EwidD-SwNIkz" }, "source": [ "次に、`'int'` ベクトル化レイヤーを使用して、1D ConvNet を構築します。" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:15.131845Z", "iopub.status.busy": "2022-12-15T01:17:15.131226Z", "iopub.status.idle": "2022-12-15T01:17:15.136020Z", "shell.execute_reply": "2022-12-15T01:17:15.135049Z" }, "id": "5ztw2XH_LbVz" }, "outputs": [], "source": [ "def create_model(vocab_size, num_labels):\n", " model = tf.keras.Sequential([\n", " layers.Embedding(vocab_size, 64, mask_zero=True),\n", " layers.Conv1D(64, 5, padding=\"valid\", activation=\"relu\", strides=2),\n", " layers.GlobalMaxPooling1D(),\n", " layers.Dense(num_labels)\n", " ])\n", " return model" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:15.139238Z", "iopub.status.busy": "2022-12-15T01:17:15.138749Z", "iopub.status.idle": "2022-12-15T01:17:20.418582Z", "shell.execute_reply": "2022-12-15T01:17:20.417683Z" }, "id": "s9rG1cFRL31Z" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/5\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 4:40 - loss: 1.3860 - accuracy: 0.2812" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 17/200 [=>............................] - ETA: 0s - loss: 1.3862 - accuracy: 0.2592 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 34/200 [====>.........................] - ETA: 0s - loss: 1.3819 - accuracy: 0.2711" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 51/200 [======>.......................] - ETA: 0s - loss: 1.3747 - accuracy: 0.3156" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 68/200 [=========>....................] - ETA: 0s - loss: 1.3647 - accuracy: 0.3488" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 85/200 [===========>..................] - ETA: 0s - loss: 1.3469 - accuracy: 0.3798" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "102/200 [==============>...............] - ETA: 0s - loss: 1.3225 - accuracy: 0.3915" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "119/200 [================>.............] - ETA: 0s - loss: 1.2936 - accuracy: 0.4178" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "135/200 [===================>..........] - ETA: 0s - loss: 1.2657 - accuracy: 0.4361" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "151/200 [=====================>........] - ETA: 0s - loss: 1.2356 - accuracy: 0.4538" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "167/200 [========================>.....] - ETA: 0s - loss: 1.2034 - accuracy: 0.4701" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "184/200 [==========================>...] - ETA: 0s - loss: 1.1677 - accuracy: 0.4895" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 2s 4ms/step - loss: 1.1424 - accuracy: 0.5008 - val_loss: 0.7508 - val_accuracy: 0.7188\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 2/5\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.9226 - accuracy: 0.6250" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 18/200 [=>............................] - ETA: 0s - loss: 0.8218 - accuracy: 0.6736" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 35/200 [====>.........................] - ETA: 0s - loss: 0.7589 - accuracy: 0.7063" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 52/200 [======>.......................] - ETA: 0s - loss: 0.7448 - accuracy: 0.7037" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 69/200 [=========>....................] - ETA: 0s - loss: 0.7279 - accuracy: 0.7074" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 86/200 [===========>..................] - ETA: 0s - loss: 0.7194 - accuracy: 0.7082" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "103/200 [==============>...............] - ETA: 0s - loss: 0.7000 - accuracy: 0.7188" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "120/200 [=================>............] - ETA: 0s - loss: 0.6862 - accuracy: 0.7292" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "137/200 [===================>..........] - ETA: 0s - loss: 0.6777 - accuracy: 0.7329" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "154/200 [======================>.......] - ETA: 0s - loss: 0.6607 - accuracy: 0.7429" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "171/200 [========================>.....] - ETA: 0s - loss: 0.6486 - accuracy: 0.7515" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "188/200 [===========================>..] - ETA: 0s - loss: 0.6339 - accuracy: 0.7580" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 4ms/step - loss: 0.6282 - accuracy: 0.7600 - val_loss: 0.5454 - val_accuracy: 0.8019\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 3/5\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.5562 - accuracy: 0.7500" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 18/200 [=>............................] - ETA: 0s - loss: 0.5219 - accuracy: 0.8264" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 35/200 [====>.........................] - ETA: 0s - loss: 0.4681 - accuracy: 0.8446" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 52/200 [======>.......................] - ETA: 0s - loss: 0.4612 - accuracy: 0.8468" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 69/200 [=========>....................] - ETA: 0s - loss: 0.4498 - accuracy: 0.8456" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 86/200 [===========>..................] - ETA: 0s - loss: 0.4450 - accuracy: 0.8503" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "103/200 [==============>...............] - ETA: 0s - loss: 0.4334 - accuracy: 0.8559" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "120/200 [=================>............] - ETA: 0s - loss: 0.4235 - accuracy: 0.8622" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "137/200 [===================>..........] - ETA: 0s - loss: 0.4153 - accuracy: 0.8650" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "155/200 [======================>.......] - ETA: 0s - loss: 0.4043 - accuracy: 0.8700" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "173/200 [========================>.....] - ETA: 0s - loss: 0.3958 - accuracy: 0.8745" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "190/200 [===========================>..] - ETA: 0s - loss: 0.3868 - accuracy: 0.8778" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 4ms/step - loss: 0.3834 - accuracy: 0.8789 - val_loss: 0.4846 - val_accuracy: 0.8150\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 4/5\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.3376 - accuracy: 0.9062" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 18/200 [=>............................] - ETA: 0s - loss: 0.3205 - accuracy: 0.9167" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 36/200 [====>.........................] - ETA: 0s - loss: 0.2761 - accuracy: 0.9271" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 53/200 [======>.......................] - ETA: 0s - loss: 0.2677 - accuracy: 0.9287" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 70/200 [=========>....................] - ETA: 0s - loss: 0.2585 - accuracy: 0.9321" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 87/200 [============>.................] - ETA: 0s - loss: 0.2564 - accuracy: 0.9343" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "104/200 [==============>...............] - ETA: 0s - loss: 0.2492 - accuracy: 0.9372" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "121/200 [=================>............] - ETA: 0s - loss: 0.2436 - accuracy: 0.9393" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "139/200 [===================>..........] - ETA: 0s - loss: 0.2373 - accuracy: 0.9415" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "156/200 [======================>.......] - ETA: 0s - loss: 0.2308 - accuracy: 0.9445" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "174/200 [=========================>....] - ETA: 0s - loss: 0.2248 - accuracy: 0.9468" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "192/200 [===========================>..] - ETA: 0s - loss: 0.2189 - accuracy: 0.9486" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 4ms/step - loss: 0.2168 - accuracy: 0.9495 - val_loss: 0.4828 - val_accuracy: 0.8188\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 5/5\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/200 [..............................] - ETA: 0s - loss: 0.1600 - accuracy: 1.0000" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 18/200 [=>............................] - ETA: 0s - loss: 0.1791 - accuracy: 0.9566" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 35/200 [====>.........................] - ETA: 0s - loss: 0.1470 - accuracy: 0.9723" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 52/200 [======>.......................] - ETA: 0s - loss: 0.1407 - accuracy: 0.9724" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 69/200 [=========>....................] - ETA: 0s - loss: 0.1343 - accuracy: 0.9751" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 86/200 [===========>..................] - ETA: 0s - loss: 0.1324 - accuracy: 0.9749" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "103/200 [==============>...............] - ETA: 0s - loss: 0.1285 - accuracy: 0.9760" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "120/200 [=================>............] - ETA: 0s - loss: 0.1247 - accuracy: 0.9771" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "137/200 [===================>..........] - ETA: 0s - loss: 0.1219 - accuracy: 0.9781" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "154/200 [======================>.......] - ETA: 0s - loss: 0.1187 - accuracy: 0.9793" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "171/200 [========================>.....] - ETA: 0s - loss: 0.1164 - accuracy: 0.9793" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "188/200 [===========================>..] - ETA: 0s - loss: 0.1131 - accuracy: 0.9799" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "200/200 [==============================] - 1s 4ms/step - loss: 0.1108 - accuracy: 0.9805 - val_loss: 0.5063 - val_accuracy: 0.8181\n" ] } ], "source": [ "# `vocab_size` is `VOCAB_SIZE + 1` since `0` is used additionally for padding.\n", "int_model = create_model(vocab_size=VOCAB_SIZE + 1, num_labels=4)\n", "int_model.compile(\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=True),\n", " optimizer='adam',\n", " metrics=['accuracy'])\n", "history = int_model.fit(int_train_ds, validation_data=int_val_ds, epochs=5)" ] }, { "cell_type": "markdown", "metadata": { "id": "x3J9Eeuv97zE" }, "source": [ "2 つのモデルを比較します。" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:20.422707Z", "iopub.status.busy": "2022-12-15T01:17:20.422144Z", "iopub.status.idle": "2022-12-15T01:17:20.432620Z", "shell.execute_reply": "2022-12-15T01:17:20.431930Z" }, "id": "N8ViDXw99v_u" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Linear model on binary vectorized data:\n", "Model: \"sequential\"\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Layer (type) Output Shape Param # \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " dense (Dense) (None, 4) 40004 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Total params: 40,004\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Trainable params: 40,004\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Non-trainable params: 0\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "None\n" ] } ], "source": [ "print(\"Linear model on binary vectorized data:\")\n", "print(binary_model.summary())" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:20.437589Z", "iopub.status.busy": "2022-12-15T01:17:20.437115Z", "iopub.status.idle": "2022-12-15T01:17:20.451011Z", "shell.execute_reply": "2022-12-15T01:17:20.450398Z" }, "id": "P9BOeoCwborD" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ConvNet model on int vectorized data:\n", "Model: \"sequential_1\"\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Layer (type) Output Shape Param # \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " embedding (Embedding) (None, None, 64) 640064 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " conv1d (Conv1D) (None, None, 64) 20544 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " global_max_pooling1d (Globa (None, 64) 0 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " lMaxPooling1D) \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " dense_1 (Dense) (None, 4) 260 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Total params: 660,868\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Trainable params: 660,868\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Non-trainable params: 0\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "None\n" ] } ], "source": [ "print(\"ConvNet model on int vectorized data:\")\n", "print(int_model.summary())" ] }, { "cell_type": "markdown", "metadata": { "id": "zYYW9tUdCtTy" }, "source": [ "テストデータで両方のモデルを評価します。" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:20.457503Z", "iopub.status.busy": "2022-12-15T01:17:20.456952Z", "iopub.status.idle": "2022-12-15T01:17:21.832601Z", "shell.execute_reply": "2022-12-15T01:17:21.831813Z" }, "id": "5dTc4nZqf7fK" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/250 [..............................] - ETA: 19s - loss: 0.5814 - accuracy: 0.8125" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 19/250 [=>............................] - ETA: 0s - loss: 0.5146 - accuracy: 0.8158 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 37/250 [===>..........................] - ETA: 0s - loss: 0.5115 - accuracy: 0.8209" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 56/250 [=====>........................] - ETA: 0s - loss: 0.5176 - accuracy: 0.8203" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 74/250 [=======>......................] - ETA: 0s - loss: 0.5223 - accuracy: 0.8180" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 92/250 [==========>...................] - ETA: 0s - loss: 0.5157 - accuracy: 0.8183" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "111/250 [============>.................] - ETA: 0s - loss: 0.5212 - accuracy: 0.8133" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "129/250 [==============>...............] - ETA: 0s - loss: 0.5149 - accuracy: 0.8173" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "147/250 [================>.............] - ETA: 0s - loss: 0.5097 - accuracy: 0.8199" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "165/250 [==================>...........] - ETA: 0s - loss: 0.5156 - accuracy: 0.8178" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "184/250 [=====================>........] - ETA: 0s - loss: 0.5152 - accuracy: 0.8173" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "202/250 [=======================>......] - ETA: 0s - loss: 0.5188 - accuracy: 0.8150" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "221/250 [=========================>....] - ETA: 0s - loss: 0.5171 - accuracy: 0.8143" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "240/250 [===========================>..] - ETA: 0s - loss: 0.5159 - accuracy: 0.8147" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "250/250 [==============================] - 1s 3ms/step - loss: 0.5173 - accuracy: 0.8148\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/250 [..............................] - ETA: 15s - loss: 0.4688 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 25/250 [==>...........................] - ETA: 0s - loss: 0.5412 - accuracy: 0.8175 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 49/250 [====>.........................] - ETA: 0s - loss: 0.5347 - accuracy: 0.8048" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 74/250 [=======>......................] - ETA: 0s - loss: 0.5274 - accuracy: 0.8045" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 98/250 [==========>...................] - ETA: 0s - loss: 0.5200 - accuracy: 0.8093" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "122/250 [=============>................] - ETA: 0s - loss: 0.5187 - accuracy: 0.8102" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "146/250 [================>.............] - ETA: 0s - loss: 0.5188 - accuracy: 0.8110" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "170/250 [===================>..........] - ETA: 0s - loss: 0.5153 - accuracy: 0.8119" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "194/250 [======================>.......] - ETA: 0s - loss: 0.5286 - accuracy: 0.8082" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "218/250 [=========================>....] - ETA: 0s - loss: 0.5318 - accuracy: 0.8048" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "242/250 [============================>.] - ETA: 0s - loss: 0.5252 - accuracy: 0.8073" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "250/250 [==============================] - 1s 2ms/step - loss: 0.5244 - accuracy: 0.8070\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Binary model accuracy: 81.48%\n", "Int model accuracy: 80.70%\n" ] } ], "source": [ "binary_loss, binary_accuracy = binary_model.evaluate(binary_test_ds)\n", "int_loss, int_accuracy = int_model.evaluate(int_test_ds)\n", "\n", "print(\"Binary model accuracy: {:2.2%}\".format(binary_accuracy))\n", "print(\"Int model accuracy: {:2.2%}\".format(int_accuracy))" ] }, { "cell_type": "markdown", "metadata": { "id": "F9dhj8Hey9DS" }, "source": [ "注意: このサンプルデータセットは、かなり単純な分類問題を表しています。より複雑なデータセットと問題は、前処理戦略とモデルアーキテクチャに微妙ながら重要な違いをもたらします。さまざまなアプローチを比較するために、さまざまなハイパーパラメータとエポックを試してみてください。" ] }, { "cell_type": "markdown", "metadata": { "id": "h9GaXTsIgP-3" }, "source": [ "### モデルをエクスポートする\n", "\n", "上記のコードでは、モデルにテキストをフィードする前に、`tf.keras.layers.TextVectorization` レイヤーをデータセットに適用しました。モデルで生の文字列を処理できるようにする場合 (たとえば、展開を簡素化するため)、モデル内に `TextVectorization` レイヤーを含めることができます。\n", "\n", "これを行うには、トレーニングしたばかりの重みを使用して新しいモデルを作成できます。" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:21.836706Z", "iopub.status.busy": "2022-12-15T01:17:21.836028Z", "iopub.status.idle": "2022-12-15T01:17:23.156544Z", "shell.execute_reply": "2022-12-15T01:17:23.155803Z" }, "id": "_bRe3KX8gRCX" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/250 [..............................] - ETA: 55s - loss: 0.4487 - accuracy: 0.9375" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 12/250 [>.............................] - ETA: 1s - loss: 0.5282 - accuracy: 0.8021 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 24/250 [=>............................] - ETA: 1s - loss: 0.5292 - accuracy: 0.8047" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 36/250 [===>..........................] - ETA: 0s - loss: 0.5131 - accuracy: 0.8177" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 48/250 [====>.........................] - ETA: 0s - loss: 0.5053 - accuracy: 0.8197" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 61/250 [======>.......................] - ETA: 0s - loss: 0.5153 - accuracy: 0.8222" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 74/250 [=======>......................] - ETA: 0s - loss: 0.5187 - accuracy: 0.8214" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 86/250 [=========>....................] - ETA: 0s - loss: 0.5156 - accuracy: 0.8198" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 99/250 [==========>...................] - ETA: 0s - loss: 0.5152 - accuracy: 0.8188" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "111/250 [============>.................] - ETA: 0s - loss: 0.5160 - accuracy: 0.8167" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "123/250 [=============>................] - ETA: 0s - loss: 0.5137 - accuracy: 0.8181" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "135/250 [===============>..............] - ETA: 0s - loss: 0.5095 - accuracy: 0.8204" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "147/250 [================>.............] - ETA: 0s - loss: 0.5104 - accuracy: 0.8191" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "159/250 [==================>...........] - ETA: 0s - loss: 0.5083 - accuracy: 0.8206" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "172/250 [===================>..........] - ETA: 0s - loss: 0.5127 - accuracy: 0.8196" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "185/250 [=====================>........] - ETA: 0s - loss: 0.5163 - accuracy: 0.8166" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "197/250 [======================>.......] - ETA: 0s - loss: 0.5170 - accuracy: 0.8158" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "209/250 [========================>.....] - ETA: 0s - loss: 0.5155 - accuracy: 0.8155" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "222/250 [=========================>....] - ETA: 0s - loss: 0.5164 - accuracy: 0.8149" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "234/250 [===========================>..] - ETA: 0s - loss: 0.5182 - accuracy: 0.8132" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "247/250 [============================>.] - ETA: 0s - loss: 0.5164 - accuracy: 0.8145" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "250/250 [==============================] - 1s 4ms/step - loss: 0.5173 - accuracy: 0.8148\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 81.48%\n" ] } ], "source": [ "export_model = tf.keras.Sequential(\n", " [binary_vectorize_layer, binary_model,\n", " layers.Activation('sigmoid')])\n", "\n", "export_model.compile(\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=False),\n", " optimizer='adam',\n", " metrics=['accuracy'])\n", "\n", "# Test it with `raw_test_ds`, which yields raw strings\n", "loss, accuracy = export_model.evaluate(raw_test_ds)\n", "print(\"Accuracy: {:2.2%}\".format(binary_accuracy))" ] }, { "cell_type": "markdown", "metadata": { "id": "m2eqTVBP4DUN" }, "source": [ "これで、モデルは生の文字列を入力として受け取り、`Model.predict` を使用して各ラベルのスコアを予測できます。最大スコアのラベルを見つける関数を定義します。" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.160442Z", "iopub.status.busy": "2022-12-15T01:17:23.159725Z", "iopub.status.idle": "2022-12-15T01:17:23.163594Z", "shell.execute_reply": "2022-12-15T01:17:23.163013Z" }, "id": "GU53uRXz45iO" }, "outputs": [], "source": [ "def get_string_labels(predicted_scores_batch):\n", " predicted_int_labels = tf.math.argmax(predicted_scores_batch, axis=1)\n", " predicted_labels = tf.gather(raw_train_ds.class_names, predicted_int_labels)\n", " return predicted_labels" ] }, { "cell_type": "markdown", "metadata": { "id": "yqnWc7Nn5eou" }, "source": [ "### 新しいデータで推論を実行する" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.167045Z", "iopub.status.busy": "2022-12-15T01:17:23.166501Z", "iopub.status.idle": "2022-12-15T01:17:23.320215Z", "shell.execute_reply": "2022-12-15T01:17:23.319396Z" }, "id": "BOR2MupW1_zS" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Question: how do I extract keys from a dict into a list?\n", "Predicted label: b'python'\n", "Question: debug public static void main(string[] args) {...}\n", "Predicted label: b'java'\n" ] } ], "source": [ "inputs = [\n", " \"how do I extract keys from a dict into a list?\", # 'python'\n", " \"debug public static void main(string[] args) {...}\", # 'java'\n", "]\n", "predicted_scores = export_model.predict(inputs)\n", "predicted_labels = get_string_labels(predicted_scores)\n", "for input, label in zip(inputs, predicted_labels):\n", " print(\"Question: \", input)\n", " print(\"Predicted label: \", label.numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "0QDVfii_4slI" }, "source": [ "モデル内にテキスト前処理ロジックを含めると、モデルを本番環境にエクスポートして展開を簡素化し、[トレーニング/テストスキュー](https://developers.google.com/machine-learning/guides/rules-of-ml#training-serving_skew)の可能性を減らすことができます。\n", "\n", "`tf.keras.layers.TextVectorization` を適用する場所を選択する際に性能の違いに留意する必要があります。モデルの外部で使用すると、GPU でトレーニングするときに非同期 CPU 処理とデータのバッファリングを行うことができます。したがって、GPU でモデルをトレーニングしている場合は、モデルの開発中に最高のパフォーマンスを得るためにこのオプションを使用し、デプロイの準備ができたらモデル内に `TextVectorization` レイヤーを含めるように切り替えることをお勧めします。\n", "\n", "モデルの保存の詳細については、[モデルの保存と読み込み](../keras/save_and_load.ipynb)チュートリアルをご覧ください。" ] }, { "cell_type": "markdown", "metadata": { "id": "p4cvuFzavTRy" }, "source": [ "## 例 2: イーリアスの翻訳者を予測する\n" ] }, { "cell_type": "markdown", "metadata": { "id": "fOlJ22508RIe" }, "source": [ "以下に、`tf.data.TextLineDataset` を使用してテキストファイルから例を読み込み、[TensorFlow Text](https://www.tensorflow.org/text) を使用してデータを前処理する例を示します。この例では、ホーマーのイーリアスの 3 つの異なる英語翻訳を使用し、与えられた 1 行のテキストから翻訳者を識別するようにモデルをトレーニングします。" ] }, { "cell_type": "markdown", "metadata": { "id": "-pCgKbOSk7kU" }, "source": [ "### データセットをダウンロードして調査する\n", "\n", "3 つのテキストの翻訳者は次のとおりです。\n", "\n", "- [ウィリアム・クーパー](https://en.wikipedia.org/wiki/William_Cowper) — [テキスト](https://storage.googleapis.com/download.tensorflow.org/data/illiad/cowper.txt)\n", "- [エドワード、ダービー伯爵](https://en.wikipedia.org/wiki/Edward_Smith-Stanley,_14th_Earl_of_Derby) — [テキスト](https://storage.googleapis.com/download.tensorflow.org/data/illiad/derby.txt)\n", "- [サミュエル・バトラー](https://en.wikipedia.org/wiki/Samuel_Butler_%28novelist%29) — [テキスト](https://storage.googleapis.com/download.tensorflow.org/data/illiad/butler.txt)\n", "\n", "このチュートリアルで使われているテキストファイルは、ヘッダ、フッタ、行番号、章のタイトルの削除など、いくつかの典型的な前処理が行われています。\n", "\n", "前処理後のファイルをローカルにダウンロードします。" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.324267Z", "iopub.status.busy": "2022-12-15T01:17:23.323516Z", "iopub.status.idle": "2022-12-15T01:17:23.482832Z", "shell.execute_reply": "2022-12-15T01:17:23.482177Z" }, "id": "4YlKQthEYlFw" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/illiad/cowper.txt\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 16384/815980 [..............................] - ETA: 0s" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "819200/815980 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "827392/815980 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/illiad/derby.txt\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 16384/809730 [..............................] - ETA: 0s" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "811008/809730 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "819200/809730 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/illiad/butler.txt\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 16384/807992 [..............................] - ETA: 0s" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "811008/807992 [==============================] - 0s 0us/step\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "819200/807992 [==============================] - 0s 0us/step\n" ] }, { "data": { "text/plain": [ "[PosixPath('/home/kbuilder/.keras/datasets/194px-New_East_River_Bridge_from_Brooklyn_det.4a09796u.jpg'),\n", " PosixPath('/home/kbuilder/.keras/datasets/spa-eng'),\n", " PosixPath('/home/kbuilder/.keras/datasets/jena_climate_2009_2016.csv'),\n", " PosixPath('/home/kbuilder/.keras/datasets/facades'),\n", " PosixPath('/home/kbuilder/.keras/datasets/mnist.npz'),\n", " PosixPath('/home/kbuilder/.keras/datasets/flower_photos.tar.gz'),\n", " PosixPath('/home/kbuilder/.keras/datasets/kandinsky5.jpg'),\n", " PosixPath('/home/kbuilder/.keras/datasets/heart.csv'),\n", " PosixPath('/home/kbuilder/.keras/datasets/ImageNetLabels.txt'),\n", " PosixPath('/home/kbuilder/.keras/datasets/jena_climate_2009_2016.csv.zip'),\n", " PosixPath('/home/kbuilder/.keras/datasets/320px-Felis_catus-cat_on_snow.jpg'),\n", " PosixPath('/home/kbuilder/.keras/datasets/Giant Panda'),\n", " PosixPath('/home/kbuilder/.keras/datasets/flower_photos'),\n", " PosixPath('/home/kbuilder/.keras/datasets/shakespeare.txt'),\n", " PosixPath('/home/kbuilder/.keras/datasets/facades.tar.gz'),\n", " PosixPath('/home/kbuilder/.keras/datasets/fashion-mnist'),\n", " PosixPath('/home/kbuilder/.keras/datasets/train.csv'),\n", " PosixPath('/home/kbuilder/.keras/datasets/derby.txt'),\n", " PosixPath('/home/kbuilder/.keras/datasets/HIGGS.csv.gz'),\n", " PosixPath('/home/kbuilder/.keras/datasets/surf.jpg'),\n", " PosixPath('/home/kbuilder/.keras/datasets/bedroom_hrnet_tutorial.jpg'),\n", " PosixPath('/home/kbuilder/.keras/datasets/spa-eng.zip'),\n", " PosixPath('/home/kbuilder/.keras/datasets/butler.txt'),\n", " PosixPath('/home/kbuilder/.keras/datasets/Fireboat'),\n", " PosixPath('/home/kbuilder/.keras/datasets/cowper.txt'),\n", " PosixPath('/home/kbuilder/.keras/datasets/YellowLabradorLooking_new.jpg')]" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "DIRECTORY_URL = 'https://storage.googleapis.com/download.tensorflow.org/data/illiad/'\n", "FILE_NAMES = ['cowper.txt', 'derby.txt', 'butler.txt']\n", "\n", "for name in FILE_NAMES:\n", " text_dir = utils.get_file(name, origin=DIRECTORY_URL + name)\n", "\n", "parent_dir = pathlib.Path(text_dir).parent\n", "list(parent_dir.iterdir())" ] }, { "cell_type": "markdown", "metadata": { "id": "M8PHK5J_cXE5" }, "source": [ "### データセットを読み込む\n", "\n", "以前は、`tf.keras.utils.text_dataset_from_directory` では、ファイルのすべてのコンテンツが 1 つの例として扱われていました。ここでは、`tf.data.TextLineDataset` を使用します。これは、テキストファイルから `tf.data.Dataset` を作成するように設計されています。それぞれの例は、元のファイルからの行です。`TextLineDataset` は、主に行ベースのテキストデータ (詩やエラーログなど) に役立ちます。\n", "\n", "これらのファイルを繰り返し処理し、各ファイルを独自のデータセットに読み込みます。各例には個別にラベルを付ける必要があるため、`Dataset.map` を使用して、それぞれにラベラー関数を適用します。これにより、データセット内のすべての例が繰り返され、 (`example, label`) ペアが返されます。" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.486414Z", "iopub.status.busy": "2022-12-15T01:17:23.485894Z", "iopub.status.idle": "2022-12-15T01:17:23.489389Z", "shell.execute_reply": "2022-12-15T01:17:23.488737Z" }, "id": "YIIWIdPXgk7I" }, "outputs": [], "source": [ "def labeler(example, index):\n", " return example, tf.cast(index, tf.int64)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.492611Z", "iopub.status.busy": "2022-12-15T01:17:23.492126Z", "iopub.status.idle": "2022-12-15T01:17:23.553800Z", "shell.execute_reply": "2022-12-15T01:17:23.553195Z" }, "id": "8Ajx7AmZnEg3" }, "outputs": [], "source": [ "labeled_data_sets = []\n", "\n", "for i, file_name in enumerate(FILE_NAMES):\n", " lines_dataset = tf.data.TextLineDataset(str(parent_dir/file_name))\n", " labeled_dataset = lines_dataset.map(lambda ex: labeler(ex, i))\n", " labeled_data_sets.append(labeled_dataset)" ] }, { "cell_type": "markdown", "metadata": { "id": "wPOsVK1e9NGM" }, "source": [ "次に、`Dataset.concatenate` を使用し、これらのラベル付きデータセットを 1 つのデータセットに結合し、`Dataset.shuffle` を使用してシャッフルします。\n" ] }, { "cell_type": "code", "execution_count": 37, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.557240Z", "iopub.status.busy": "2022-12-15T01:17:23.556964Z", "iopub.status.idle": "2022-12-15T01:17:23.560354Z", "shell.execute_reply": "2022-12-15T01:17:23.559752Z" }, "id": "6jAeYkTIi9-2" }, "outputs": [], "source": [ "BUFFER_SIZE = 50000\n", "BATCH_SIZE = 64\n", "VALIDATION_SIZE = 5000" ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.563443Z", "iopub.status.busy": "2022-12-15T01:17:23.563005Z", "iopub.status.idle": "2022-12-15T01:17:23.567725Z", "shell.execute_reply": "2022-12-15T01:17:23.567123Z" }, "id": "Qd544E-Sh63L" }, "outputs": [], "source": [ "all_labeled_data = labeled_data_sets[0]\n", "for labeled_dataset in labeled_data_sets[1:]:\n", " all_labeled_data = all_labeled_data.concatenate(labeled_dataset)\n", "\n", "all_labeled_data = all_labeled_data.shuffle(\n", " BUFFER_SIZE, reshuffle_each_iteration=False)" ] }, { "cell_type": "markdown", "metadata": { "id": "r4JEHrJXeG5k" }, "source": [ "前述の手順でいくつかの例を出力します。データセットはまだバッチ処理されていないため、`all_labeled_data` の各エントリは 1 つのデータポイントに対応します。" ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:23.571302Z", "iopub.status.busy": "2022-12-15T01:17:23.570793Z", "iopub.status.idle": "2022-12-15T01:17:24.166936Z", "shell.execute_reply": "2022-12-15T01:17:24.166172Z" }, "id": "gywKlN0xh6u5" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Sentence: b\"the man's right shoulder, and then stuck in the ground. He stood stock\"\n", "Label: 2\n", "Sentence: b'For yet a child he left me, when he fell'\n", "Label: 1\n", "Sentence: b'Save me, my brother! Pity me! Thy steeds'\n", "Label: 0\n", "Sentence: b'prepared the mess she bade them drink it. When they had done so and had'\n", "Label: 2\n", "Sentence: b'exhorts him to return to the field of battle. An interview succeeds'\n", "Label: 0\n", "Sentence: b'Then said Achilles, \"Son of Atreus, king of men Agamemnon, see to these'\n", "Label: 2\n", "Sentence: b'The hand of Menelaus, and while all'\n", "Label: 0\n", "Sentence: b\"A huge bull's hide, all drench'd and soak'd with grease;\"\n", "Label: 1\n", "Sentence: b'He sat, where sat the other Powers divine,'\n", "Label: 0\n", "Sentence: b'And in my cause lies slain, of any Greek'\n", "Label: 1\n" ] } ], "source": [ "for text, label in all_labeled_data.take(10):\n", " print(\"Sentence: \", text.numpy())\n", " print(\"Label:\", label.numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "5rrpU2_sfDh0" }, "source": [ "### トレーニング用データセットを準備する\n", "\n", "`tf.keras.layers.TextVectorization` を使用してテキストデータセットを前処理する代わりに、TensorFlow Text API を使用してデータを標準化およびトークン化し、語彙を作成し、`tf.lookup.StaticVocabularyTable` を使用してトークンを整数にマッピングし、モデルにフィードします。(詳細については [TensorFlow Text](https://www.tensorflow.org/text) を参照してください)。\n", "\n", "テキストを小文字に変換してトークン化する関数を定義します。\n", "\n", "- TensorFlow Text は、さまざまなトークナイザーを提供します。この例では、`text.UnicodeScriptTokenizer` を使用してデータセットをトークン化します。\n", "- `Dataset.map` を使用して、トークン化をデータセットに適用します。" ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:24.170805Z", "iopub.status.busy": "2022-12-15T01:17:24.170167Z", "iopub.status.idle": "2022-12-15T01:17:24.173997Z", "shell.execute_reply": "2022-12-15T01:17:24.173374Z" }, "id": "v4DpQW-Y12rm" }, "outputs": [], "source": [ "tokenizer = tf_text.UnicodeScriptTokenizer()" ] }, { "cell_type": "code", "execution_count": 41, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:24.177276Z", "iopub.status.busy": "2022-12-15T01:17:24.176653Z", "iopub.status.idle": "2022-12-15T01:17:24.180317Z", "shell.execute_reply": "2022-12-15T01:17:24.179608Z" }, "id": "pz8xEj0ugu51" }, "outputs": [], "source": [ "def tokenize(text, unused_label):\n", " lower_case = tf_text.case_fold_utf8(text)\n", " return tokenizer.tokenize(lower_case)" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:24.183434Z", "iopub.status.busy": "2022-12-15T01:17:24.182861Z", "iopub.status.idle": "2022-12-15T01:17:25.828121Z", "shell.execute_reply": "2022-12-15T01:17:25.827293Z" }, "id": "vzUrAzOq31QL" }, "outputs": [], "source": [ "tokenized_ds = all_labeled_data.map(tokenize)" ] }, { "cell_type": "markdown", "metadata": { "id": "jx4Q2i8XLV7o" }, "source": [ "データセットを反復処理して、トークン化されたいくつかの例を出力します。\n" ] }, { "cell_type": "code", "execution_count": 43, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:25.832378Z", "iopub.status.busy": "2022-12-15T01:17:25.832105Z", "iopub.status.idle": "2022-12-15T01:17:27.035425Z", "shell.execute_reply": "2022-12-15T01:17:27.034687Z" }, "id": "g2mkWri7LiGq" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Tokens: [b'the' b'man' b\"'\" b's' b'right' b'shoulder' b',' b'and' b'then' b'stuck'\n", " b'in' b'the' b'ground' b'.' b'he' b'stood' b'stock']\n", "Tokens: [b'for' b'yet' b'a' b'child' b'he' b'left' b'me' b',' b'when' b'he'\n", " b'fell']\n", "Tokens: [b'save' b'me' b',' b'my' b'brother' b'!' b'pity' b'me' b'!' b'thy'\n", " b'steeds']\n", "Tokens: [b'prepared' b'the' b'mess' b'she' b'bade' b'them' b'drink' b'it' b'.'\n", " b'when' b'they' b'had' b'done' b'so' b'and' b'had']\n", "Tokens: [b'exhorts' b'him' b'to' b'return' b'to' b'the' b'field' b'of' b'battle'\n", " b'.' b'an' b'interview' b'succeeds']\n" ] } ], "source": [ "for text_batch in tokenized_ds.take(5):\n", " print(\"Tokens: \", text_batch.numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "JPd4PsskJ_Xt" }, "source": [ "次に、トークンを頻度で並べ替え、上位の `VOCAB_SIZE` トークンを保持することにより、語彙を構築します。" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:27.039376Z", "iopub.status.busy": "2022-12-15T01:17:27.038645Z", "iopub.status.idle": "2022-12-15T01:17:34.170689Z", "shell.execute_reply": "2022-12-15T01:17:34.169939Z" }, "id": "YkHtbGnDh6mg" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Vocab size: 10000\n", "First five vocab entries: [b',', b'the', b'and', b\"'\", b'of']\n" ] } ], "source": [ "tokenized_ds = configure_dataset(tokenized_ds)\n", "\n", "vocab_dict = collections.defaultdict(lambda: 0)\n", "for toks in tokenized_ds.as_numpy_iterator():\n", " for tok in toks:\n", " vocab_dict[tok] += 1\n", "\n", "vocab = sorted(vocab_dict.items(), key=lambda x: x[1], reverse=True)\n", "vocab = [token for token, count in vocab]\n", "vocab = vocab[:VOCAB_SIZE]\n", "vocab_size = len(vocab)\n", "print(\"Vocab size: \", vocab_size)\n", "print(\"First five vocab entries:\", vocab[:5])" ] }, { "cell_type": "markdown", "metadata": { "id": "PyKSsaNAKi17" }, "source": [ "トークンを整数に変換するには、`vocab` セットを使用して、`tf.lookup.StaticVocabularyTable` を作成します。トークンを [`2`, `vocab_size + 2`] の範囲の整数にマップします。`TextVectorization` レイヤーと同様に、`0` はパディングを示すために予約されており、`1` は語彙外 (OOV) トークンを示すために予約されています。" ] }, { "cell_type": "code", "execution_count": 45, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:34.174700Z", "iopub.status.busy": "2022-12-15T01:17:34.174101Z", "iopub.status.idle": "2022-12-15T01:17:34.182839Z", "shell.execute_reply": "2022-12-15T01:17:34.182192Z" }, "id": "kCBo2yFHD7y6" }, "outputs": [], "source": [ "keys = vocab\n", "values = range(2, len(vocab) + 2) # Reserve `0` for padding, `1` for OOV tokens.\n", "\n", "init = tf.lookup.KeyValueTensorInitializer(\n", " keys, values, key_dtype=tf.string, value_dtype=tf.int64)\n", "\n", "num_oov_buckets = 1\n", "vocab_table = tf.lookup.StaticVocabularyTable(init, num_oov_buckets)" ] }, { "cell_type": "markdown", "metadata": { "id": "Z5F-EiBpOADE" }, "source": [ "最後に、トークナイザーとルックアップテーブルを使用して、データセットを標準化、トークン化、およびベクトル化する関数を定義します。" ] }, { "cell_type": "code", "execution_count": 46, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:34.186317Z", "iopub.status.busy": "2022-12-15T01:17:34.185794Z", "iopub.status.idle": "2022-12-15T01:17:34.189528Z", "shell.execute_reply": "2022-12-15T01:17:34.188880Z" }, "id": "HcIQ7LOTh6eT" }, "outputs": [], "source": [ "def preprocess_text(text, label):\n", " standardized = tf_text.case_fold_utf8(text)\n", " tokenized = tokenizer.tokenize(standardized)\n", " vectorized = vocab_table.lookup(tokenized)\n", " return vectorized, label" ] }, { "cell_type": "markdown", "metadata": { "id": "v6S5Qyabi-vo" }, "source": [ "1 つの例でこれを試して、出力を確認します。" ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:34.192603Z", "iopub.status.busy": "2022-12-15T01:17:34.192364Z", "iopub.status.idle": "2022-12-15T01:17:34.850703Z", "shell.execute_reply": "2022-12-15T01:17:34.849918Z" }, "id": "jgxPZaxUuTbk" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Sentence: b\"the man's right shoulder, and then stuck in the ground. He stood stock\"\n", "Vectorized sentence: [ 3 86 5 29 274 527 2 4 33 2749 13 3 191 7\n", " 12 108 3286]\n" ] } ], "source": [ "example_text, example_label = next(iter(all_labeled_data))\n", "print(\"Sentence: \", example_text.numpy())\n", "vectorized_text, example_label = preprocess_text(example_text, example_label)\n", "print(\"Vectorized sentence: \", vectorized_text.numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "p9qHM0v8k_Mg" }, "source": [ "次に、`Dataset.map` を使用して、データセットに対して前処理関数を実行します。" ] }, { "cell_type": "code", "execution_count": 48, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:34.854313Z", "iopub.status.busy": "2022-12-15T01:17:34.853763Z", "iopub.status.idle": "2022-12-15T01:17:36.357711Z", "shell.execute_reply": "2022-12-15T01:17:36.356976Z" }, "id": "KmQVsAgJ-RM0" }, "outputs": [], "source": [ "all_encoded_data = all_labeled_data.map(preprocess_text)" ] }, { "cell_type": "markdown", "metadata": { "id": "_YZToSXSm0qr" }, "source": [ "### データセットをトレーニング用セットとテスト用セットに分割する\n" ] }, { "cell_type": "markdown", "metadata": { "id": "itxIJwkrUXgv" }, "source": [ "Keras `TextVectorization` レイヤーでも、ベクトル化されたデータをバッチ処理してパディングします。バッチ内の例は同じサイズと形状である必要があるため、パディングが必要です。これらのデータセットの例はすべて同じサイズではありません。テキストの各行には、異なる数の単語があります。\n", "\n", "`tf.data.Dataset` は、データセットの分割とパディングのバッチ処理をサポートしています" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:36.362290Z", "iopub.status.busy": "2022-12-15T01:17:36.362000Z", "iopub.status.idle": "2022-12-15T01:17:36.367115Z", "shell.execute_reply": "2022-12-15T01:17:36.366519Z" }, "id": "r-rmbijQh6bf" }, "outputs": [], "source": [ "train_data = all_encoded_data.skip(VALIDATION_SIZE).shuffle(BUFFER_SIZE)\n", "validation_data = all_encoded_data.take(VALIDATION_SIZE)" ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:36.370516Z", "iopub.status.busy": "2022-12-15T01:17:36.369917Z", "iopub.status.idle": "2022-12-15T01:17:36.377447Z", "shell.execute_reply": "2022-12-15T01:17:36.376799Z" }, "id": "qTP0IwHBCn0Q" }, "outputs": [], "source": [ "train_data = train_data.padded_batch(BATCH_SIZE)\n", "validation_data = validation_data.padded_batch(BATCH_SIZE)" ] }, { "cell_type": "markdown", "metadata": { "id": "m-wmFq8uW1zS" }, "source": [ "`validation_data` および `train_data` は (`example, label`) ペアのコレクションではなく、バッチのコレクションです。各バッチは、配列として表される (*多くの例*、*多くのラベル*) のペアです。\n", "\n", "以下に示します。" ] }, { "cell_type": "code", "execution_count": 51, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:36.380785Z", "iopub.status.busy": "2022-12-15T01:17:36.380255Z", "iopub.status.idle": "2022-12-15T01:17:37.569554Z", "shell.execute_reply": "2022-12-15T01:17:37.568785Z" }, "id": "kMslWfuwoqpB" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Text batch shape: (64, 19)\n", "Label batch shape: (64,)\n", "First text example: tf.Tensor(\n", "[ 3 86 5 29 274 527 2 4 33 2749 13 3 191 7\n", " 12 108 3286 0 0], shape=(19,), dtype=int64)\n", "First label example: tf.Tensor(2, shape=(), dtype=int64)\n" ] } ], "source": [ "sample_text, sample_labels = next(iter(validation_data))\n", "print(\"Text batch shape: \", sample_text.shape)\n", "print(\"Label batch shape: \", sample_labels.shape)\n", "print(\"First text example: \", sample_text[0])\n", "print(\"First label example: \", sample_labels[0])" ] }, { "cell_type": "markdown", "metadata": { "id": "UI4I6_Sa0vWu" }, "source": [ "パディングに 0 を使用し、語彙外 (OOV) トークンに 1 を使用するため、語彙のサイズが 2 つ増えました。" ] }, { "cell_type": "code", "execution_count": 52, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:37.573307Z", "iopub.status.busy": "2022-12-15T01:17:37.572735Z", "iopub.status.idle": "2022-12-15T01:17:37.575981Z", "shell.execute_reply": "2022-12-15T01:17:37.575341Z" }, "id": "u21LlkO8QGRX" }, "outputs": [], "source": [ "vocab_size += 2" ] }, { "cell_type": "markdown", "metadata": { "id": "h44Ox11OYLP-" }, "source": [ "以前と同じように、パフォーマンスを向上させるためにデータセットを構成します。" ] }, { "cell_type": "code", "execution_count": 53, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:37.579289Z", "iopub.status.busy": "2022-12-15T01:17:37.578747Z", "iopub.status.idle": "2022-12-15T01:17:37.583608Z", "shell.execute_reply": "2022-12-15T01:17:37.583043Z" }, "id": "BpT0b_7mYRXV" }, "outputs": [], "source": [ "train_data = configure_dataset(train_data)\n", "validation_data = configure_dataset(validation_data)" ] }, { "cell_type": "markdown", "metadata": { "id": "K8SUhGFNsmRi" }, "source": [ "### モデルをトレーニングする\n", "\n", "以前と同じように、このデータセットでモデルをトレーニングできます。" ] }, { "cell_type": "code", "execution_count": 54, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:17:37.586997Z", "iopub.status.busy": "2022-12-15T01:17:37.586430Z", "iopub.status.idle": "2022-12-15T01:18:12.383152Z", "shell.execute_reply": "2022-12-15T01:18:12.382344Z" }, "id": "QJgI1pow2YR9" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/Unknown - 22s 22s/step - loss: 1.0935 - accuracy: 0.4844" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 15/Unknown - 22s 4ms/step - loss: 1.0591 - accuracy: 0.4292" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 32/Unknown - 22s 3ms/step - loss: 1.0370 - accuracy: 0.4038" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 51/Unknown - 22s 3ms/step - loss: 1.0025 - accuracy: 0.4577" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 70/Unknown - 22s 3ms/step - loss: 0.9636 - accuracy: 0.4975" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 88/Unknown - 22s 3ms/step - loss: 0.9299 - accuracy: 0.5234" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 107/Unknown - 22s 3ms/step - loss: 0.8895 - accuracy: 0.5510" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 125/Unknown - 22s 3ms/step - loss: 0.8618 - accuracy: 0.5709" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 144/Unknown - 22s 3ms/step - loss: 0.8311 - accuracy: 0.5885" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 162/Unknown - 23s 3ms/step - loss: 0.8055 - accuracy: 0.6055" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 181/Unknown - 23s 3ms/step - loss: 0.7815 - accuracy: 0.6205" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 200/Unknown - 23s 3ms/step - loss: 0.7567 - accuracy: 0.6363" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 219/Unknown - 23s 3ms/step - loss: 0.7330 - accuracy: 0.6493" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 238/Unknown - 23s 3ms/step - loss: 0.7147 - accuracy: 0.6595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 257/Unknown - 23s 3ms/step - loss: 0.6987 - accuracy: 0.6685" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 276/Unknown - 23s 3ms/step - loss: 0.6831 - accuracy: 0.6769" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 295/Unknown - 23s 3ms/step - loss: 0.6682 - accuracy: 0.6856" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 314/Unknown - 23s 3ms/step - loss: 0.6548 - accuracy: 0.6934" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 333/Unknown - 23s 3ms/step - loss: 0.6429 - accuracy: 0.6998" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 351/Unknown - 23s 3ms/step - loss: 0.6332 - accuracy: 0.7053" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 369/Unknown - 23s 3ms/step - loss: 0.6250 - accuracy: 0.7096" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 387/Unknown - 23s 3ms/step - loss: 0.6158 - accuracy: 0.7150" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 406/Unknown - 23s 3ms/step - loss: 0.6065 - accuracy: 0.7203" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 425/Unknown - 23s 3ms/step - loss: 0.5976 - accuracy: 0.7256" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 444/Unknown - 23s 3ms/step - loss: 0.5900 - accuracy: 0.7300" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 463/Unknown - 23s 3ms/step - loss: 0.5825 - accuracy: 0.7334" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 482/Unknown - 23s 3ms/step - loss: 0.5752 - accuracy: 0.7372" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 501/Unknown - 23s 3ms/step - loss: 0.5680 - accuracy: 0.7407" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 520/Unknown - 24s 3ms/step - loss: 0.5617 - accuracy: 0.7443" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 539/Unknown - 24s 3ms/step - loss: 0.5558 - accuracy: 0.7474" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 558/Unknown - 24s 3ms/step - loss: 0.5505 - accuracy: 0.7501" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 576/Unknown - 24s 3ms/step - loss: 0.5451 - accuracy: 0.7530" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 594/Unknown - 24s 3ms/step - loss: 0.5398 - accuracy: 0.7560" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 612/Unknown - 24s 3ms/step - loss: 0.5349 - accuracy: 0.7587" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 630/Unknown - 24s 3ms/step - loss: 0.5306 - accuracy: 0.7612" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 648/Unknown - 24s 3ms/step - loss: 0.5263 - accuracy: 0.7633" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 667/Unknown - 24s 3ms/step - loss: 0.5222 - accuracy: 0.7654" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 685/Unknown - 24s 3ms/step - loss: 0.5187 - accuracy: 0.7673" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "697/697 [==============================] - 28s 9ms/step - loss: 0.5163 - accuracy: 0.7686 - val_loss: 0.3744 - val_accuracy: 0.8406\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 2/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/697 [..............................] - ETA: 4s - loss: 0.3166 - accuracy: 0.8750" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 19/697 [..............................] - ETA: 1s - loss: 0.3328 - accuracy: 0.8635" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 38/697 [>.............................] - ETA: 1s - loss: 0.3559 - accuracy: 0.8532" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 57/697 [=>............................] - ETA: 1s - loss: 0.3428 - accuracy: 0.8586" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 76/697 [==>...........................] - ETA: 1s - loss: 0.3381 - accuracy: 0.8588" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 95/697 [===>..........................] - ETA: 1s - loss: 0.3365 - accuracy: 0.8602" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "114/697 [===>..........................] - ETA: 1s - loss: 0.3339 - accuracy: 0.8601" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "132/697 [====>.........................] - ETA: 1s - loss: 0.3345 - accuracy: 0.8604" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "151/697 [=====>........................] - ETA: 1s - loss: 0.3309 - accuracy: 0.8624" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "169/697 [======>.......................] - ETA: 1s - loss: 0.3268 - accuracy: 0.8646" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "188/697 [=======>......................] - ETA: 1s - loss: 0.3229 - accuracy: 0.8657" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "207/697 [=======>......................] - ETA: 1s - loss: 0.3173 - accuracy: 0.8677" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "226/697 [========>.....................] - ETA: 1s - loss: 0.3110 - accuracy: 0.8711" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "245/697 [=========>....................] - ETA: 1s - loss: 0.3082 - accuracy: 0.8724" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "264/697 [==========>...................] - ETA: 1s - loss: 0.3074 - accuracy: 0.8730" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "283/697 [===========>..................] - ETA: 1s - loss: 0.3046 - accuracy: 0.8741" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "302/697 [===========>..................] - ETA: 1s - loss: 0.3029 - accuracy: 0.8753" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "321/697 [============>.................] - ETA: 1s - loss: 0.3005 - accuracy: 0.8765" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "340/697 [=============>................] - ETA: 0s - loss: 0.2990 - accuracy: 0.8772" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "359/697 [==============>...............] - ETA: 0s - loss: 0.2982 - accuracy: 0.8780" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "378/697 [===============>..............] - ETA: 0s - loss: 0.2972 - accuracy: 0.8779" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "397/697 [================>.............] - ETA: 0s - loss: 0.2954 - accuracy: 0.8783" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "416/697 [================>.............] - ETA: 0s - loss: 0.2935 - accuracy: 0.8795" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "435/697 [=================>............] - ETA: 0s - loss: 0.2917 - accuracy: 0.8806" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "454/697 [==================>...........] - ETA: 0s - loss: 0.2906 - accuracy: 0.8809" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "473/697 [===================>..........] - ETA: 0s - loss: 0.2890 - accuracy: 0.8814" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "492/697 [====================>.........] - ETA: 0s - loss: 0.2882 - accuracy: 0.8817" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "511/697 [====================>.........] - ETA: 0s - loss: 0.2863 - accuracy: 0.8826" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "530/697 [=====================>........] - ETA: 0s - loss: 0.2859 - accuracy: 0.8830" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "549/697 [======================>.......] - ETA: 0s - loss: 0.2849 - accuracy: 0.8835" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "568/697 [=======================>......] - ETA: 0s - loss: 0.2840 - accuracy: 0.8843" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "587/697 [========================>.....] - ETA: 0s - loss: 0.2830 - accuracy: 0.8850" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "606/697 [=========================>....] - ETA: 0s - loss: 0.2814 - accuracy: 0.8858" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "625/697 [=========================>....] - ETA: 0s - loss: 0.2805 - accuracy: 0.8864" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "644/697 [==========================>...] - ETA: 0s - loss: 0.2795 - accuracy: 0.8865" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "663/697 [===========================>..] - ETA: 0s - loss: 0.2786 - accuracy: 0.8871" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "682/697 [============================>.] - ETA: 0s - loss: 0.2782 - accuracy: 0.8870" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "697/697 [==============================] - 3s 4ms/step - loss: 0.2777 - accuracy: 0.8872 - val_loss: 0.3690 - val_accuracy: 0.8472\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 3/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/697 [..............................] - ETA: 3s - loss: 0.2072 - accuracy: 0.9375" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 19/697 [..............................] - ETA: 1s - loss: 0.2354 - accuracy: 0.9071" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 38/697 [>.............................] - ETA: 1s - loss: 0.2411 - accuracy: 0.9062" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 57/697 [=>............................] - ETA: 1s - loss: 0.2305 - accuracy: 0.9109" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 76/697 [==>...........................] - ETA: 1s - loss: 0.2265 - accuracy: 0.9130" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 94/697 [===>..........................] - ETA: 1s - loss: 0.2268 - accuracy: 0.9136" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "112/697 [===>..........................] - ETA: 1s - loss: 0.2258 - accuracy: 0.9129" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "131/697 [====>.........................] - ETA: 1s - loss: 0.2272 - accuracy: 0.9116" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "150/697 [=====>........................] - ETA: 1s - loss: 0.2246 - accuracy: 0.9124" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "169/697 [======>.......................] - ETA: 1s - loss: 0.2207 - accuracy: 0.9146" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "188/697 [=======>......................] - ETA: 1s - loss: 0.2192 - accuracy: 0.9145" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "206/697 [=======>......................] - ETA: 1s - loss: 0.2153 - accuracy: 0.9166" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "224/697 [========>.....................] - ETA: 1s - loss: 0.2110 - accuracy: 0.9185" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "242/697 [=========>....................] - ETA: 1s - loss: 0.2085 - accuracy: 0.9199" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "260/697 [==========>...................] - ETA: 1s - loss: 0.2076 - accuracy: 0.9204" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "279/697 [===========>..................] - ETA: 1s - loss: 0.2058 - accuracy: 0.9212" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "298/697 [===========>..................] - ETA: 1s - loss: 0.2044 - accuracy: 0.9221" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "317/697 [============>.................] - ETA: 1s - loss: 0.2032 - accuracy: 0.9226" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "336/697 [=============>................] - ETA: 1s - loss: 0.2031 - accuracy: 0.9222" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "355/697 [==============>...............] - ETA: 0s - loss: 0.2017 - accuracy: 0.9228" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "374/697 [===============>..............] - ETA: 0s - loss: 0.2013 - accuracy: 0.9228" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "392/697 [===============>..............] - ETA: 0s - loss: 0.2007 - accuracy: 0.9227" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "410/697 [================>.............] - ETA: 0s - loss: 0.1992 - accuracy: 0.9232" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "428/697 [=================>............] - ETA: 0s - loss: 0.1982 - accuracy: 0.9238" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "447/697 [==================>...........] - ETA: 0s - loss: 0.1971 - accuracy: 0.9245" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "465/697 [===================>..........] - ETA: 0s - loss: 0.1964 - accuracy: 0.9245" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "484/697 [===================>..........] - ETA: 0s - loss: 0.1953 - accuracy: 0.9251" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "503/697 [====================>.........] - ETA: 0s - loss: 0.1936 - accuracy: 0.9257" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "522/697 [=====================>........] - ETA: 0s - loss: 0.1927 - accuracy: 0.9263" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "541/697 [======================>.......] - ETA: 0s - loss: 0.1920 - accuracy: 0.9266" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "559/697 [=======================>......] - ETA: 0s - loss: 0.1916 - accuracy: 0.9267" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "578/697 [=======================>......] - ETA: 0s - loss: 0.1913 - accuracy: 0.9269" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "596/697 [========================>.....] - ETA: 0s - loss: 0.1901 - accuracy: 0.9275" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "615/697 [=========================>....] - ETA: 0s - loss: 0.1893 - accuracy: 0.9278" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "634/697 [==========================>...] - ETA: 0s - loss: 0.1886 - accuracy: 0.9283" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "653/697 [===========================>..] - ETA: 0s - loss: 0.1878 - accuracy: 0.9286" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "672/697 [===========================>..] - ETA: 0s - loss: 0.1868 - accuracy: 0.9292" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "690/697 [============================>.] - ETA: 0s - loss: 0.1865 - accuracy: 0.9294" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "697/697 [==============================] - 3s 4ms/step - loss: 0.1862 - accuracy: 0.9295 - val_loss: 0.4152 - val_accuracy: 0.8382\n" ] } ], "source": [ "model = create_model(vocab_size=vocab_size, num_labels=3)\n", "\n", "model.compile(\n", " optimizer='adam',\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=True),\n", " metrics=['accuracy'])\n", "\n", "history = model.fit(train_data, validation_data=validation_data, epochs=3)" ] }, { "cell_type": "code", "execution_count": 55, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:12.387426Z", "iopub.status.busy": "2022-12-15T01:18:12.386807Z", "iopub.status.idle": "2022-12-15T01:18:13.173767Z", "shell.execute_reply": "2022-12-15T01:18:13.173029Z" }, "id": "KTPCYf_Jh6TH" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/Unknown - 1s 603ms/step - loss: 0.4481 - accuracy: 0.8438" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 27/Unknown - 1s 2ms/step - loss: 0.3972 - accuracy: 0.8414 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 54/Unknown - 1s 2ms/step - loss: 0.4058 - accuracy: 0.8409" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "79/79 [==============================] - 1s 2ms/step - loss: 0.4152 - accuracy: 0.8382\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Loss: 0.41523823142051697\n", "Accuracy: 83.82%\n" ] } ], "source": [ "loss, accuracy = model.evaluate(validation_data)\n", "\n", "print(\"Loss: \", loss)\n", "print(\"Accuracy: {:2.2%}\".format(accuracy))" ] }, { "cell_type": "markdown", "metadata": { "id": "_knIsO-r4pHb" }, "source": [ "### モデルをエクスポートする" ] }, { "cell_type": "markdown", "metadata": { "id": "FEuMLJA_Xiwo" }, "source": [ "モデルが生の文字列を入力として受け取ることができるようにするには、カスタム前処理関数と同じ手順を実行する `TextVectorization` レイヤーを作成します。すでに語彙をトレーニングしているので、新しい語彙をトレーニングする `TextVectorization.adapt` の代わりに、`TextVectorization.set_vocabulary` を使用できます。" ] }, { "cell_type": "code", "execution_count": 56, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:13.177644Z", "iopub.status.busy": "2022-12-15T01:18:13.177010Z", "iopub.status.idle": "2022-12-15T01:18:13.192126Z", "shell.execute_reply": "2022-12-15T01:18:13.191420Z" }, "id": "_ODkRXbk6aHb" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/tmpfs/src/tf_docs_env/lib/python3.9/site-packages/numpy/core/numeric.py:2468: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n", " return bool(asarray(a1 == a2).all())\n", "/tmpfs/src/tf_docs_env/lib/python3.9/site-packages/keras/layers/preprocessing/index_lookup.py:458: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n", " if self.mask_token is not None and self.mask_token in tokens:\n" ] } ], "source": [ "preprocess_layer = TextVectorization(\n", " max_tokens=vocab_size,\n", " standardize=tf_text.case_fold_utf8,\n", " split=tokenizer.tokenize,\n", " output_mode='int',\n", " output_sequence_length=MAX_SEQUENCE_LENGTH)\n", "\n", "preprocess_layer.set_vocabulary(vocab)" ] }, { "cell_type": "code", "execution_count": 57, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:13.195162Z", "iopub.status.busy": "2022-12-15T01:18:13.194936Z", "iopub.status.idle": "2022-12-15T01:18:13.205623Z", "shell.execute_reply": "2022-12-15T01:18:13.205050Z" }, "id": "G-Cvd27y4qwt" }, "outputs": [], "source": [ "export_model = tf.keras.Sequential(\n", " [preprocess_layer, model,\n", " layers.Activation('sigmoid')])\n", "\n", "export_model.compile(\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=False),\n", " optimizer='adam',\n", " metrics=['accuracy'])" ] }, { "cell_type": "code", "execution_count": 58, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:13.208807Z", "iopub.status.busy": "2022-12-15T01:18:13.208300Z", "iopub.status.idle": "2022-12-15T01:18:18.964906Z", "shell.execute_reply": "2022-12-15T01:18:18.964155Z" }, "id": "Pyg0B4zsc-UD" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2022-12-15 01:18:17.427116: W tensorflow/core/grappler/optimizers/loop_optimizer.cc:907] Skipping loop optimization for Merge node with control input: sequential_4/text_vectorization_2/UnicodeScriptTokenize/Assert_1/AssertGuard/branch_executed/_185\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/Unknown - 5s 5s/step - loss: 0.7075 - accuracy: 0.7344" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 8/Unknown - 5s 7ms/step - loss: 0.5362 - accuracy: 0.7734" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 16/Unknown - 5s 7ms/step - loss: 0.5461 - accuracy: 0.7852" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 24/Unknown - 5s 7ms/step - loss: 0.5640 - accuracy: 0.7773" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 32/Unknown - 5s 7ms/step - loss: 0.5520 - accuracy: 0.7788" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 40/Unknown - 5s 7ms/step - loss: 0.5366 - accuracy: 0.7863" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 48/Unknown - 6s 7ms/step - loss: 0.5563 - accuracy: 0.7806" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 56/Unknown - 6s 7ms/step - loss: 0.5648 - accuracy: 0.7799" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 64/Unknown - 6s 7ms/step - loss: 0.5714 - accuracy: 0.7786" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 71/Unknown - 6s 7ms/step - loss: 0.5681 - accuracy: 0.7779" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 78/Unknown - 6s 7ms/step - loss: 0.5733 - accuracy: 0.7764" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "79/79 [==============================] - 6s 7ms/step - loss: 0.5736 - accuracy: 0.7764\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Loss: 0.5736407041549683\n", "Accuracy: 77.64%\n" ] } ], "source": [ "# Create a test dataset of raw strings.\n", "test_ds = all_labeled_data.take(VALIDATION_SIZE).batch(BATCH_SIZE)\n", "test_ds = configure_dataset(test_ds)\n", "\n", "loss, accuracy = export_model.evaluate(test_ds)\n", "\n", "print(\"Loss: \", loss)\n", "print(\"Accuracy: {:2.2%}\".format(accuracy))" ] }, { "cell_type": "markdown", "metadata": { "id": "o6Mm0Y9QYQwE" }, "source": [ "エンコードされた検証セットのモデルと生の検証セットのエクスポートされたモデルの損失と正確度は、予想どおり同じです。" ] }, { "cell_type": "markdown", "metadata": { "id": "Stk2BP8GE-qo" }, "source": [ "### 新しいデータで推論を実行する" ] }, { "cell_type": "code", "execution_count": 59, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:18.968257Z", "iopub.status.busy": "2022-12-15T01:18:18.967996Z", "iopub.status.idle": "2022-12-15T01:18:21.146388Z", "shell.execute_reply": "2022-12-15T01:18:21.145674Z" }, "id": "-w1fQGJPD2Yh" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2022-12-15 01:18:20.841752: W tensorflow/core/grappler/optimizers/loop_optimizer.cc:907] Skipping loop optimization for Merge node with control input: sequential_4/text_vectorization_2/UnicodeScriptTokenize/Assert_1/AssertGuard/branch_executed/_185\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Question: Join'd to th' Ionians with their flowing robes,\n", "Predicted label: 1\n", "Question: the allies, and his armour flashed about him so that he seemed to all\n", "Predicted label: 2\n", "Question: And with loud clangor of his arms he fell.\n", "Predicted label: 0\n" ] } ], "source": [ "inputs = [\n", " \"Join'd to th' Ionians with their flowing robes,\", # Label: 1\n", " \"the allies, and his armour flashed about him so that he seemed to all\", # Label: 2\n", " \"And with loud clangor of his arms he fell.\", # Label: 0\n", "]\n", "\n", "predicted_scores = export_model.predict(inputs)\n", "predicted_labels = tf.math.argmax(predicted_scores, axis=1)\n", "\n", "for input, label in zip(inputs, predicted_labels):\n", " print(\"Question: \", input)\n", " print(\"Predicted label: \", label.numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "9eA8TVdnA-3L" }, "source": [ "## TensorFlow Datasets (TFDS) を使用して、より多くのデータセットをダウンロードする\n" ] }, { "cell_type": "markdown", "metadata": { "id": "2QFSxfZ3Vqsn" }, "source": [ "[TensorFlow Dataset](https://www.tensorflow.org/datasets/catalog/overview) からより多くのデータセットをダウンロードできます。\n", "\n", "この例では、[IMDB 大規模映画レビューデータセット](https://www.tensorflow.org/datasets/catalog/imdb_reviews)を使用して、感情分類のモデルをトレーニングします。" ] }, { "cell_type": "code", "execution_count": 60, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:21.150267Z", "iopub.status.busy": "2022-12-15T01:18:21.149707Z", "iopub.status.idle": "2022-12-15T01:18:21.872059Z", "shell.execute_reply": "2022-12-15T01:18:21.871371Z" }, "id": "NzC65LOaVw0B" }, "outputs": [], "source": [ "# Training set.\n", "train_ds = tfds.load(\n", " 'imdb_reviews',\n", " split='train[:80%]',\n", " batch_size=BATCH_SIZE,\n", " shuffle_files=True,\n", " as_supervised=True)" ] }, { "cell_type": "code", "execution_count": 61, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:21.876193Z", "iopub.status.busy": "2022-12-15T01:18:21.875764Z", "iopub.status.idle": "2022-12-15T01:18:22.400650Z", "shell.execute_reply": "2022-12-15T01:18:22.399906Z" }, "id": "XKGkgPBkFh0k" }, "outputs": [], "source": [ "# Validation set.\n", "val_ds = tfds.load(\n", " 'imdb_reviews',\n", " split='train[80%:]',\n", " batch_size=BATCH_SIZE,\n", " shuffle_files=True,\n", " as_supervised=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "BQjf3YZAb5Ne" }, "source": [ "いくつかの例を出力します。" ] }, { "cell_type": "code", "execution_count": 62, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:22.405048Z", "iopub.status.busy": "2022-12-15T01:18:22.404476Z", "iopub.status.idle": "2022-12-15T01:18:22.751552Z", "shell.execute_reply": "2022-12-15T01:18:22.750693Z" }, "id": "Bq1w8MnfWt2C" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Review: b\"Instead, go to the zoo, buy some peanuts and feed 'em to the monkeys. Monkeys are funny. People with amnesia who don't say much, just sit there with vacant eyes are not all that funny.

Black comedy? There isn't a black person in it, and there isn't one funny thing in it either.

Walmart buys these things up somehow and puts them on their dollar rack. It's labeled Unrated. I think they took out the topless scene. They may have taken out other stuff too, who knows? All we know is that whatever they took out, isn't there any more.

The acting seemed OK to me. There's a lot of unfathomables tho. It's supposed to be a city? It's supposed to be a big lake? If it's so hot in the church people are fanning themselves, why are they all wearing coats?\"\n", "Label: 0\n", "Review: b'Well, was Morgan Freeman any more unusual as God than George Burns? This film sure was better than that bore, \"Oh, God\". I was totally engrossed and LMAO all the way through. Carrey was perfect as the out of sorts anchorman wannabe, and Aniston carried off her part as the frustrated girlfriend in her usual well played performance. I, for one, don\\'t consider her to be either ugly or untalented. I think my favorite scene was when Carrey opened up the file cabinet thinking it could never hold his life history. See if you can spot the file in the cabinet that holds the events of his bathroom humor: I was rolling over this one. Well written and even better played out, this comedy will go down as one of this funnyman\\'s best.'\n", "Label: 1\n", "Review: b'I remember stumbling upon this special while channel-surfing in 1965. I had never heard of Barbra before. When the show was over, I thought \"This is probably the best thing on TV I will ever see in my life.\" 42 years later, that has held true. There is still nothing so amazing, so honestly astonishing as the talent that was displayed here. You can talk about all the super-stars you want to, this is the most superlative of them all!

You name it, she can do it. Comedy, pathos, sultry seduction, ballads, Barbra is truly a story-teller. Her ability to pull off anything she attempts is legendary. But this special was made in the beginning, and helped to create the legend that she quickly became. In spite of rising so far in such a short time, she has fulfilled the promise, revealing more of her talents as she went along. But they are all here from the very beginning. You will not be disappointed in viewing this.'\n", "Label: 1\n", "Review: b\"Firstly, I would like to point out that people who have criticised this film have made some glaring errors. Anything that has a rating below 6/10 is clearly utter nonsense.

Creep is an absolutely fantastic film with amazing film effects. The actors are highly believable, the narrative thought provoking and the horror and graphical content extremely disturbing.

There is much mystique in this film. Many questions arise as the audience are revealed to the strange and freakish creature that makes habitat in the dark rat ridden tunnels. How was 'Craig' created and what happened to him?

A fantastic film with a large chill factor. A film with so many unanswered questions and a film that needs to be appreciated along with others like 28 Days Later, The Bunker, Dog Soldiers and Deathwatch.

Look forward to more of these fantastic films!!\"\n", "Label: 1\n", "Review: b\"I'm sorry but I didn't like this doc very much. I can think of a million ways it could have been better. The people who made it obviously don't have much imagination. The interviews aren't very interesting and no real insight is offered. The footage isn't assembled in a very informative way, either. It's too bad because this is a movie that really deserves spellbinding special features. One thing I'll say is that Isabella Rosselini gets more beautiful the older she gets. All considered, this only gets a '4.'\"\n", "Label: 0\n" ] } ], "source": [ "for review_batch, label_batch in val_ds.take(1):\n", " for i in range(5):\n", " print(\"Review: \", review_batch[i].numpy())\n", " print(\"Label: \", label_batch[i].numpy())" ] }, { "cell_type": "markdown", "metadata": { "id": "q-lVaukyb75k" }, "source": [ "これで、以前と同じようにデータを前処理してモデルをトレーニングできます。\n", "\n", "注意: これは二項分類の問題であるため、モデルには `tf.keras.losses.SparseCategoricalCrossentropy` の代わりに `tf.keras.losses.BinaryCrossentropy` を使用します。" ] }, { "cell_type": "markdown", "metadata": { "id": "ciz2CxAsZw3Z" }, "source": [ "### トレーニング用データセットを準備する" ] }, { "cell_type": "code", "execution_count": 63, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:22.755791Z", "iopub.status.busy": "2022-12-15T01:18:22.755067Z", "iopub.status.idle": "2022-12-15T01:18:24.862788Z", "shell.execute_reply": "2022-12-15T01:18:24.862038Z" }, "id": "UzT_t9ihZLH4" }, "outputs": [], "source": [ "vectorize_layer = TextVectorization(\n", " max_tokens=VOCAB_SIZE,\n", " output_mode='int',\n", " output_sequence_length=MAX_SEQUENCE_LENGTH)\n", "\n", "# Make a text-only dataset (without labels), then call `TextVectorization.adapt`.\n", "train_text = train_ds.map(lambda text, labels: text)\n", "vectorize_layer.adapt(train_text)" ] }, { "cell_type": "code", "execution_count": 64, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:24.867249Z", "iopub.status.busy": "2022-12-15T01:18:24.866716Z", "iopub.status.idle": "2022-12-15T01:18:24.870472Z", "shell.execute_reply": "2022-12-15T01:18:24.869882Z" }, "id": "zz-Xrd_ZZ4tB" }, "outputs": [], "source": [ "def vectorize_text(text, label):\n", " text = tf.expand_dims(text, -1)\n", " return vectorize_layer(text), label" ] }, { "cell_type": "code", "execution_count": 65, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:24.873541Z", "iopub.status.busy": "2022-12-15T01:18:24.873316Z", "iopub.status.idle": "2022-12-15T01:18:24.958614Z", "shell.execute_reply": "2022-12-15T01:18:24.957724Z" }, "id": "ycn0Itd6g5aF" }, "outputs": [], "source": [ "train_ds = train_ds.map(vectorize_text)\n", "val_ds = val_ds.map(vectorize_text)" ] }, { "cell_type": "code", "execution_count": 66, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:24.962452Z", "iopub.status.busy": "2022-12-15T01:18:24.961939Z", "iopub.status.idle": "2022-12-15T01:18:24.966768Z", "shell.execute_reply": "2022-12-15T01:18:24.966133Z" }, "id": "jc11jQTlZ5lj" }, "outputs": [], "source": [ "# Configure datasets for performance as before.\n", "train_ds = configure_dataset(train_ds)\n", "val_ds = configure_dataset(val_ds)" ] }, { "cell_type": "markdown", "metadata": { "id": "SQzoYkaGZ82Z" }, "source": [ "### モデルを作成、構成、およびトレーニングする" ] }, { "cell_type": "code", "execution_count": 67, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:24.970212Z", "iopub.status.busy": "2022-12-15T01:18:24.969739Z", "iopub.status.idle": "2022-12-15T01:18:25.008721Z", "shell.execute_reply": "2022-12-15T01:18:25.008048Z" }, "id": "B9IOTLkyZ-a7" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"sequential_5\"\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " Layer (type) Output Shape Param # \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " embedding_2 (Embedding) (None, None, 64) 640064 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " conv1d_2 (Conv1D) (None, None, 64) 20544 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " global_max_pooling1d_2 (Glo (None, 64) 0 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " balMaxPooling1D) \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " dense_3 (Dense) (None, 1) 65 \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " \n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "=================================================================\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Total params: 660,673\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Trainable params: 660,673\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Non-trainable params: 0\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "_________________________________________________________________\n" ] } ], "source": [ "model = create_model(vocab_size=VOCAB_SIZE + 1, num_labels=1)\n", "model.summary()" ] }, { "cell_type": "code", "execution_count": 68, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:25.015538Z", "iopub.status.busy": "2022-12-15T01:18:25.015069Z", "iopub.status.idle": "2022-12-15T01:18:25.022755Z", "shell.execute_reply": "2022-12-15T01:18:25.022124Z" }, "id": "xLnDs5dhaBAk" }, "outputs": [], "source": [ "model.compile(\n", " loss=losses.BinaryCrossentropy(from_logits=True),\n", " optimizer='adam',\n", " metrics=['accuracy'])" ] }, { "cell_type": "code", "execution_count": 69, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:25.026218Z", "iopub.status.busy": "2022-12-15T01:18:25.025768Z", "iopub.status.idle": "2022-12-15T01:18:35.083804Z", "shell.execute_reply": "2022-12-15T01:18:35.082993Z" }, "id": "rq59QpNzaDMa" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/313 [..............................] - ETA: 28:06 - loss: 0.6936 - accuracy: 0.5156" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 14/313 [>.............................] - ETA: 1s - loss: 0.6935 - accuracy: 0.4866 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 28/313 [=>............................] - ETA: 1s - loss: 0.6923 - accuracy: 0.4944" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 43/313 [===>..........................] - ETA: 0s - loss: 0.6911 - accuracy: 0.5011" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 58/313 [====>.........................] - ETA: 0s - loss: 0.6893 - accuracy: 0.4984" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 73/313 [=====>........................] - ETA: 0s - loss: 0.6872 - accuracy: 0.4929" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 87/313 [=======>......................] - ETA: 0s - loss: 0.6844 - accuracy: 0.4960" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 98/313 [========>.....................] - ETA: 0s - loss: 0.6808 - accuracy: 0.4984" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "110/313 [=========>....................] - ETA: 0s - loss: 0.6756 - accuracy: 0.4963" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "122/313 [==========>...................] - ETA: 0s - loss: 0.6695 - accuracy: 0.5029" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "135/313 [===========>..................] - ETA: 0s - loss: 0.6611 - accuracy: 0.5096" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "147/313 [=============>................] - ETA: 0s - loss: 0.6522 - accuracy: 0.5216" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "159/313 [==============>...............] - ETA: 0s - loss: 0.6406 - accuracy: 0.5419" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "172/313 [===============>..............] - ETA: 0s - loss: 0.6305 - accuracy: 0.5571" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "185/313 [================>.............] - ETA: 0s - loss: 0.6211 - accuracy: 0.5709" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "199/313 [==================>...........] - ETA: 0s - loss: 0.6093 - accuracy: 0.5852" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "212/313 [===================>..........] - ETA: 0s - loss: 0.6001 - accuracy: 0.5976" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "224/313 [====================>.........] - ETA: 0s - loss: 0.5895 - accuracy: 0.6092" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "237/313 [=====================>........] - ETA: 0s - loss: 0.5801 - accuracy: 0.6204" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "249/313 [======================>.......] - ETA: 0s - loss: 0.5722 - accuracy: 0.6291" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "261/313 [========================>.....] - ETA: 0s - loss: 0.5637 - accuracy: 0.6383" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "273/313 [=========================>....] - ETA: 0s - loss: 0.5573 - accuracy: 0.6450" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "286/313 [==========================>...] - ETA: 0s - loss: 0.5507 - accuracy: 0.6523" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "299/313 [===========================>..] - ETA: 0s - loss: 0.5438 - accuracy: 0.6593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "312/313 [============================>.] - ETA: 0s - loss: 0.5377 - accuracy: 0.6662" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "313/313 [==============================] - 7s 6ms/step - loss: 0.5375 - accuracy: 0.6665 - val_loss: 0.3738 - val_accuracy: 0.8266\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 2/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/313 [..............................] - ETA: 1s - loss: 0.5018 - accuracy: 0.7344" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 16/313 [>.............................] - ETA: 1s - loss: 0.3671 - accuracy: 0.8301" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 31/313 [=>............................] - ETA: 0s - loss: 0.3781 - accuracy: 0.8266" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 46/313 [===>..........................] - ETA: 0s - loss: 0.3673 - accuracy: 0.8325" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 61/313 [====>.........................] - ETA: 0s - loss: 0.3670 - accuracy: 0.8327" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 76/313 [======>.......................] - ETA: 0s - loss: 0.3659 - accuracy: 0.8347" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 91/313 [=======>......................] - ETA: 0s - loss: 0.3610 - accuracy: 0.8360" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "106/313 [=========>....................] - ETA: 0s - loss: 0.3543 - accuracy: 0.8395" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "121/313 [==========>...................] - ETA: 0s - loss: 0.3535 - accuracy: 0.8397" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "136/313 [============>.................] - ETA: 0s - loss: 0.3486 - accuracy: 0.8424" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "151/313 [=============>................] - ETA: 0s - loss: 0.3418 - accuracy: 0.8469" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "166/313 [==============>...............] - ETA: 0s - loss: 0.3397 - accuracy: 0.8478" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "181/313 [================>.............] - ETA: 0s - loss: 0.3359 - accuracy: 0.8496" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "196/313 [=================>............] - ETA: 0s - loss: 0.3323 - accuracy: 0.8526" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "211/313 [===================>..........] - ETA: 0s - loss: 0.3285 - accuracy: 0.8545" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "226/313 [====================>.........] - ETA: 0s - loss: 0.3227 - accuracy: 0.8572" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "241/313 [======================>.......] - ETA: 0s - loss: 0.3181 - accuracy: 0.8601" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "256/313 [=======================>......] - ETA: 0s - loss: 0.3148 - accuracy: 0.8619" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "271/313 [========================>.....] - ETA: 0s - loss: 0.3124 - accuracy: 0.8628" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "286/313 [==========================>...] - ETA: 0s - loss: 0.3088 - accuracy: 0.8646" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "301/313 [===========================>..] - ETA: 0s - loss: 0.3052 - accuracy: 0.8666" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "313/313 [==============================] - 1s 4ms/step - loss: 0.3038 - accuracy: 0.8673 - val_loss: 0.3151 - val_accuracy: 0.8632\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Epoch 3/3\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/313 [..............................] - ETA: 1s - loss: 0.3158 - accuracy: 0.8281" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 16/313 [>.............................] - ETA: 1s - loss: 0.2253 - accuracy: 0.9033" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 31/313 [=>............................] - ETA: 0s - loss: 0.2355 - accuracy: 0.9007" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 46/313 [===>..........................] - ETA: 0s - loss: 0.2269 - accuracy: 0.9056" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 61/313 [====>.........................] - ETA: 0s - loss: 0.2296 - accuracy: 0.9034" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 76/313 [======>.......................] - ETA: 0s - loss: 0.2301 - accuracy: 0.9054" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 91/313 [=======>......................] - ETA: 0s - loss: 0.2282 - accuracy: 0.9061" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "106/313 [=========>....................] - ETA: 0s - loss: 0.2221 - accuracy: 0.9098" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "121/313 [==========>...................] - ETA: 0s - loss: 0.2216 - accuracy: 0.9101" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "136/313 [============>.................] - ETA: 0s - loss: 0.2186 - accuracy: 0.9118" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "151/313 [=============>................] - ETA: 0s - loss: 0.2137 - accuracy: 0.9145" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "166/313 [==============>...............] - ETA: 0s - loss: 0.2123 - accuracy: 0.9159" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "181/313 [================>.............] - ETA: 0s - loss: 0.2096 - accuracy: 0.9166" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "196/313 [=================>............] - ETA: 0s - loss: 0.2077 - accuracy: 0.9176" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "211/313 [===================>..........] - ETA: 0s - loss: 0.2051 - accuracy: 0.9187" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "226/313 [====================>.........] - ETA: 0s - loss: 0.2011 - accuracy: 0.9208" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "241/313 [======================>.......] - ETA: 0s - loss: 0.1984 - accuracy: 0.9225" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "256/313 [=======================>......] - ETA: 0s - loss: 0.1963 - accuracy: 0.9237" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "271/313 [========================>.....] - ETA: 0s - loss: 0.1947 - accuracy: 0.9240" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "286/313 [==========================>...] - ETA: 0s - loss: 0.1921 - accuracy: 0.9254" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "301/313 [===========================>..] - ETA: 0s - loss: 0.1894 - accuracy: 0.9270" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "313/313 [==============================] - 1s 4ms/step - loss: 0.1885 - accuracy: 0.9276 - val_loss: 0.3167 - val_accuracy: 0.8624\n" ] } ], "source": [ "history = model.fit(train_ds, validation_data=val_ds, epochs=3)" ] }, { "cell_type": "code", "execution_count": 70, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:35.087574Z", "iopub.status.busy": "2022-12-15T01:18:35.087019Z", "iopub.status.idle": "2022-12-15T01:18:35.289390Z", "shell.execute_reply": "2022-12-15T01:18:35.288675Z" }, "id": "gCMWCEtyaEbR" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\r", " 1/79 [..............................] - ETA: 3s - loss: 0.2385 - accuracy: 0.8906" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "27/79 [=========>....................] - ETA: 0s - loss: 0.3132 - accuracy: 0.8663" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "53/79 [===================>..........] - ETA: 0s - loss: 0.3166 - accuracy: 0.8611" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", "79/79 [==============================] - 0s 2ms/step - loss: 0.3167 - accuracy: 0.8624\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Loss: 0.3166657090187073\n", "Accuracy: 86.24%\n" ] } ], "source": [ "loss, accuracy = model.evaluate(val_ds)\n", "\n", "print(\"Loss: \", loss)\n", "print(\"Accuracy: {:2.2%}\".format(accuracy))" ] }, { "cell_type": "markdown", "metadata": { "id": "jGtqLXVnaaFy" }, "source": [ "### モデルをエクスポートする" ] }, { "cell_type": "code", "execution_count": 71, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:35.292837Z", "iopub.status.busy": "2022-12-15T01:18:35.292246Z", "iopub.status.idle": "2022-12-15T01:18:35.339401Z", "shell.execute_reply": "2022-12-15T01:18:35.338791Z" }, "id": "yE9WZARZaZr1" }, "outputs": [], "source": [ "export_model = tf.keras.Sequential(\n", " [vectorize_layer, model,\n", " layers.Activation('sigmoid')])\n", "\n", "export_model.compile(\n", " loss=losses.SparseCategoricalCrossentropy(from_logits=False),\n", " optimizer='adam',\n", " metrics=['accuracy'])" ] }, { "cell_type": "code", "execution_count": 72, "metadata": { "execution": { "iopub.execute_input": "2022-12-15T01:18:35.342877Z", "iopub.status.busy": "2022-12-15T01:18:35.342372Z", "iopub.status.idle": "2022-12-15T01:18:35.476336Z", "shell.execute_reply": "2022-12-15T01:18:35.475570Z" }, "id": "bhF8tDH-afoC" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Question: This is a fantastic movie.\n", "Predicted label: 1\n", "Question: This is a bad movie.\n", "Predicted label: 0\n", "Question: This movie was so bad that it was good.\n", "Predicted label: 0\n", "Question: I will never say yes to watching this movie.\n", "Predicted label: 1\n" ] } ], "source": [ "# 0 --> negative review\n", "# 1 --> positive review\n", "inputs = [\n", " \"This is a fantastic movie.\",\n", " \"This is a bad movie.\",\n", " \"This movie was so bad that it was good.\",\n", " \"I will never say yes to watching this movie.\",\n", "]\n", "\n", "predicted_scores = export_model.predict(inputs)\n", "predicted_labels = [int(round(x[0])) for x in predicted_scores]\n", "\n", "for input, label in zip(inputs, predicted_labels):\n", " print(\"Question: \", input)\n", " print(\"Predicted label: \", label)" ] }, { "cell_type": "markdown", "metadata": { "id": "q1KSXDFPWiPN" }, "source": [ "## まとめ\n", "\n", "このチュートリアルでは、テキストを読み込んで前処理するいくつかの方法を示しました。次のステップとして、以下のような [TensorFlow Text](https://www.tensorflow.org/text) テキスト前処理 チュートリアルをご覧ください。\n", "\n", "- [TF テキストを使用した BERT 前処理](https://www.tensorflow.org/text/guide/bert_preprocessing_guide)\n", "- [TF Text を使用したトークン化](https://www.tensorflow.org/text/guide/tokenizers)\n", "- [サブワードトークナイザー](https://www.tensorflow.org/text/guide/subwords_tokenizer)\n", "\n", "[TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/overview) でも新しいデータセットを見つけることができます。また、`tf.data` の詳細については、[入力パイプラインの構築](../../guide/data.ipynb)に関するガイドをご覧ください。" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "text.ipynb", "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "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": "3.9.16" } }, "nbformat": 4, "nbformat_minor": 0 }