{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "Tce3stUlHN0L" }, "source": [ "##### Copyright 2019 The TensorFlow Authors.\n" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "cellView": "form", "execution": { "iopub.execute_input": "2024-01-11T18:16:47.674110Z", "iopub.status.busy": "2024-01-11T18:16:47.673724Z", "iopub.status.idle": "2024-01-11T18:16:47.677368Z", "shell.execute_reply": "2024-01-11T18:16:47.676776Z" }, "id": "tuOe1ymfHZPu" }, "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": "MT-LkFOl2axM" }, "source": [ "# Keras で DTensors を使用する" ] }, { "cell_type": "markdown", "metadata": { "id": "r6P32iYYV27b" }, "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": "vTe9dcbUAwqx" }, "source": [ "## 概要\n", "\n", "このチュートリアルでは、Keras で DTensor を使用する方法について学習します。\n", "\n", "DTensor を Keras と組み合わせることで、分散型機械学習モデルの構築とトレーニングに既存の Keras レイヤーとモデルを再利用することができます。\n", "\n", "MNIST データを使用してマルチレイヤーの分類モデルをトレーニングします。サブクラス化モデル、Sequential モデル、Functional モデルのレイアウトの設定について説明します。\n", "\n", "このチュートリアルでは、すでに「[DTensor プログラミングガイド](/guide/dtensor_overview)」を読んでいること、`Mesh` や `Layout` などの基本的な DTensor の概念に精通していることを前提としています。\n", "\n", "このチュートリアルでは、https://www.tensorflow.org/datasets/keras_example を基盤に使用しています。" ] }, { "cell_type": "markdown", "metadata": { "id": "keIyP3IoA1o4" }, "source": [ "## MNIST モデルをビルドする\n", "\n", "DTensor は、TensorFlow 2.9.0 リリースに含まれています。" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:47.681061Z", "iopub.status.busy": "2024-01-11T18:16:47.680626Z", "iopub.status.idle": "2024-01-11T18:16:49.923315Z", "shell.execute_reply": "2024-01-11T18:16:49.922380Z" }, "id": "4dHik7NYA5vm" }, "outputs": [], "source": [ "!pip install --quiet --upgrade --pre tensorflow tensorflow-datasets" ] }, { "cell_type": "markdown", "metadata": { "id": "VttBMZngDx8x" }, "source": [ "次に、`tensorflow` と `tensorflow.experimental.dtensor` をインポートし、8 個の仮想 CPU を使用するように TensorFlow を構成します。\n", "\n", "この例では CPU を使用しますが、DTensor は CPU、GPU、または TPU デバイスで同じように動作します。" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:49.928280Z", "iopub.status.busy": "2024-01-11T18:16:49.927598Z", "iopub.status.idle": "2024-01-11T18:16:52.587184Z", "shell.execute_reply": "2024-01-11T18:16:52.586484Z" }, "id": "CodX6idGBGSm" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2024-01-11 18:16:50.357646: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n", "2024-01-11 18:16:50.357693: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n", "2024-01-11 18:16:50.359187: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n" ] } ], "source": [ "import tensorflow as tf\n", "import tensorflow_datasets as tfds\n", "from tensorflow.experimental import dtensor" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:52.591608Z", "iopub.status.busy": "2024-01-11T18:16:52.590812Z", "iopub.status.idle": "2024-01-11T18:16:54.542465Z", "shell.execute_reply": "2024-01-11T18:16:54.541401Z" }, "id": "aAtvrpasDpDD" }, "outputs": [], "source": [ "def configure_virtual_cpus(ncpu):\n", " phy_devices = tf.config.list_physical_devices('CPU')\n", " tf.config.set_logical_device_configuration(\n", " phy_devices[0], \n", " [tf.config.LogicalDeviceConfiguration()] * ncpu)\n", " \n", "configure_virtual_cpus(8)\n", "tf.config.list_logical_devices('CPU')\n", "\n", "devices = [f'CPU:{i}' for i in range(8)]" ] }, { "cell_type": "markdown", "metadata": { "id": "ogULE1OHtyd9" }, "source": [ "## 決定論的疑似乱数ジェネレータ\n", "\n", "1 つ注意しておかなければならないのは、DTensor API では、実行中の各クライアントに同じランダムシードがある必要があることです。そうすることで、重みの初期化で決定論的動作が得られます。これは、Keras で `tf.keras.utils.set_random_seed()` を使ってグローバルシードを設定することで行えます。" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.546312Z", "iopub.status.busy": "2024-01-11T18:16:54.545783Z", "iopub.status.idle": "2024-01-11T18:16:54.562732Z", "shell.execute_reply": "2024-01-11T18:16:54.561841Z" }, "id": "9u85YypguL8N" }, "outputs": [], "source": [ "tf.keras.backend.experimental.enable_tf_random_generator()\n", "tf.keras.utils.set_random_seed(1337)" ] }, { "cell_type": "markdown", "metadata": { "id": "tO11XvPDAu3_" }, "source": [ "## データ並列メッシュを作成する\n", "\n", "このチュートリアルでは、データ並列トレーニングを実演します。モデルの並列トレーニングと空間の並列トレーニングへの適応は、別のセットの `Layout` オブジェクトに切り替えるのと同じくらい単純です。データ並列を超える分散トレーニングについての詳細は、[DTensor の詳細な ML チュートリアル](https://www.tensorflow.org/tutorials/distribute/dtensor_ml_tutorial)をご覧ください。\n", "\n", "データ並列トレーニングは、一般的に使用される並列トレーニングですが、`tf.distribute.MirroredStrategy` などによっても使用されます。\n", "\n", "DTensor を使うと、データ並列トレーニングループは、単一の 'batch' 次元で構成される `Mesh` を使用します。各デバイスは、グローバルの batch からシャードを受け取るモデルのレプリカを実行します。\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.566185Z", "iopub.status.busy": "2024-01-11T18:16:54.565948Z", "iopub.status.idle": "2024-01-11T18:16:54.570271Z", "shell.execute_reply": "2024-01-11T18:16:54.569406Z" }, "id": "6sT6s6z4j9H-" }, "outputs": [], "source": [ "mesh = dtensor.create_mesh([(\"batch\", 8)], devices=devices)" ] }, { "cell_type": "markdown", "metadata": { "id": "rouFcF6FE0aF" }, "source": [ "各デバイスがモデルの完全なレプリカを実行する過程で、モデルの変数がメッシュ（シャーディングなし）間で完全に複製されます。例として、この `Mesh` の階数 2 の重みに対して完全に複製されるレイアウトは、以下のようになります。" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.574008Z", "iopub.status.busy": "2024-01-11T18:16:54.573530Z", "iopub.status.idle": "2024-01-11T18:16:54.577385Z", "shell.execute_reply": "2024-01-11T18:16:54.576572Z" }, "id": "U8OxvkDKE1Nu" }, "outputs": [], "source": [ "example_weight_layout = dtensor.Layout([dtensor.UNSHARDED, dtensor.UNSHARDED], mesh) # or\n", "example_weight_layout = dtensor.Layout.replicated(mesh, rank=2)" ] }, { "cell_type": "markdown", "metadata": { "id": "6Bnic98RE0xi" }, "source": [ "階数 2 のデータテンソルのレイアウトは、最初の次元（`batch_sharded` としても知られます）に沿ってシャーディングされます。" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.580717Z", "iopub.status.busy": "2024-01-11T18:16:54.580276Z", "iopub.status.idle": "2024-01-11T18:16:54.584329Z", "shell.execute_reply": "2024-01-11T18:16:54.583541Z" }, "id": "PhYp0EKBFfxt" }, "outputs": [], "source": [ "example_data_layout = dtensor.Layout(['batch', dtensor.UNSHARDED], mesh) # or\n", "example_data_layout = dtensor.Layout.batch_sharded(mesh, 'batch', rank=2)" ] }, { "cell_type": "markdown", "metadata": { "id": "4U-6n0DericV" }, "source": [ "## レイアウトで Keras レイヤーを作成する\n", "\n", "データ並列スキームでは通常、モデルの各レプリカがシャーディングされた入力データを使って計算を行えるように、完全に複製されたレイアウトを使ってモデルの重みを作成します。\n", "\n", "レイヤーの重みのレイアウト情報を構成するために、Keras では、ほとんどの組み込みレイヤーで使用できる追加のパラメータをレイヤーコンストラクタに公開しています。\n", "\n", "以下は、完全に複製された重みレイアウトを使用する小さな画像分類モデルを構築する例です。レイアウト情報の `kernel` と `bias` は、`kernel_layout` と `bias_layout` の引数を介して `tf.keras.layers.Dense` に指定できます。組み込み Keras レイヤーのほとんどは、レイアウトの重みの `Layout` を明示的に指定できるようになっています。" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.587860Z", "iopub.status.busy": "2024-01-11T18:16:54.587391Z", "iopub.status.idle": "2024-01-11T18:16:54.591035Z", "shell.execute_reply": "2024-01-11T18:16:54.590247Z" }, "id": "Koc5GlA1tFXY" }, "outputs": [], "source": [ "unsharded_layout_2d = dtensor.Layout.replicated(mesh, 2)\n", "unsharded_layout_1d = dtensor.Layout.replicated(mesh, 1)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:54.594179Z", "iopub.status.busy": "2024-01-11T18:16:54.593694Z", "iopub.status.idle": "2024-01-11T18:16:55.040851Z", "shell.execute_reply": "2024-01-11T18:16:55.040127Z" }, "id": "GfOGTIxGs5Ql" }, "outputs": [], "source": [ "model = tf.keras.models.Sequential([\n", " tf.keras.layers.Flatten(input_shape=(28, 28)),\n", " tf.keras.layers.Dense(128, \n", " activation='relu',\n", " name='d1',\n", " kernel_layout=unsharded_layout_2d, \n", " bias_layout=unsharded_layout_1d),\n", " tf.keras.layers.Dense(10,\n", " name='d2',\n", " kernel_layout=unsharded_layout_2d, \n", " bias_layout=unsharded_layout_1d)\n", "])" ] }, { "cell_type": "markdown", "metadata": { "id": "0frf3jsVtx_n" }, "source": [ "レイアウト情報は、重みの `layout` プロパティを調べることで確認できます。" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:55.044455Z", "iopub.status.busy": "2024-01-11T18:16:55.044211Z", "iopub.status.idle": "2024-01-11T18:16:55.047955Z", "shell.execute_reply": "2024-01-11T18:16:55.047378Z" }, "id": "Z_nqv_VdwcXo" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Weight name: d1/kernel:0 with layout: Layout.from_string(sharding_specs:unsharded,unsharded, mesh:|batch=8|0,1,2,3,4,5,6,7|0,1,2,3,4,5,6,7|/job:localhost/replica:0/task:0/device:CPU:0,/job:localhost/replica:0/task:0/device:CPU:1,/job:localhost/replica:0/task:0/device:CPU:2,/job:localhost/replica:0/task:0/device:CPU:3,/job:localhost/replica:0/task:0/device:CPU:4,/job:localhost/replica:0/task:0/device:CPU:5,/job:localhost/replica:0/task:0/device:CPU:6,/job:localhost/replica:0/task:0/device:CPU:7)\n" ] } ], "source": [ "for weight in model.weights:\n", " print(f'Weight name: {weight.name} with layout: {weight.layout}')\n", " break" ] }, { "cell_type": "markdown", "metadata": { "id": "6FMGB-QsxPtU" }, "source": [ "## データセットを読み込んで入力パイプラインを構築する\n", "\n", "MNIST データセットを読み込んで、それに使用する事前処理用の入力パイプラインを構成します。データセット自体は DTensor レイアウト情報に関連付けられていません。今後の TensorFlow リリースにおいて、`tf.data` との DTensor Keras 統合が改善される予定です。\n" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:55.050814Z", "iopub.status.busy": "2024-01-11T18:16:55.050551Z", "iopub.status.idle": "2024-01-11T18:16:57.609464Z", "shell.execute_reply": "2024-01-11T18:16:57.608733Z" }, "id": "zGt4kwltxOt4" }, "outputs": [], "source": [ "(ds_train, ds_test), ds_info = tfds.load(\n", " 'mnist',\n", " split=['train', 'test'],\n", " shuffle_files=True,\n", " as_supervised=True,\n", " with_info=True,\n", ")" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.613610Z", "iopub.status.busy": "2024-01-11T18:16:57.612991Z", "iopub.status.idle": "2024-01-11T18:16:57.616855Z", "shell.execute_reply": "2024-01-11T18:16:57.616276Z" }, "id": "HkUaOB_ryaLH" }, "outputs": [], "source": [ "def normalize_img(image, label):\n", " \"\"\"Normalizes images: `uint8` -> `float32`.\"\"\"\n", " return tf.cast(image, tf.float32) / 255., label" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.619938Z", "iopub.status.busy": "2024-01-11T18:16:57.619722Z", "iopub.status.idle": "2024-01-11T18:16:57.662251Z", "shell.execute_reply": "2024-01-11T18:16:57.661567Z" }, "id": "Efm2H1iqydan" }, "outputs": [], "source": [ "batch_size = 128\n", "\n", "ds_train = ds_train.map(\n", " normalize_img, num_parallel_calls=tf.data.AUTOTUNE)\n", "ds_train = ds_train.cache()\n", "ds_train = ds_train.shuffle(ds_info.splits['train'].num_examples)\n", "ds_train = ds_train.batch(batch_size)\n", "ds_train = ds_train.prefetch(tf.data.AUTOTUNE)" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.665516Z", "iopub.status.busy": "2024-01-11T18:16:57.665270Z", "iopub.status.idle": "2024-01-11T18:16:57.680167Z", "shell.execute_reply": "2024-01-11T18:16:57.679602Z" }, "id": "Lcrg6QAtyis4" }, "outputs": [], "source": [ "ds_test = ds_test.map(\n", " normalize_img, num_parallel_calls=tf.data.AUTOTUNE)\n", "ds_test = ds_test.batch(batch_size)\n", "ds_test = ds_test.cache()\n", "ds_test = ds_test.prefetch(tf.data.AUTOTUNE)" ] }, { "cell_type": "markdown", "metadata": { "id": "fHEZwib7lhqn" }, "source": [ "## モデルのトレーニングロジックを定義する\n", "\n", "次に、モデルのトレーニングロジックと評価ロジックを定義します。\n", "\n", "TensorFlow 2.9 の時点では、DTensor が有効化された Keras モデルにカスタムトレーニングループを書き込む必要があります。これは、入力データに、Keras の標準の `tf.keras.Model.fit()` または `tf.keras.Model.eval()` 関数には組み込まれていない適切なレイアウト情報を詰め込むために行います。さらに多くの `tf.data` サポートが、今後のリリースで追加される予定です。 " ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.683699Z", "iopub.status.busy": "2024-01-11T18:16:57.683472Z", "iopub.status.idle": "2024-01-11T18:16:57.688508Z", "shell.execute_reply": "2024-01-11T18:16:57.687919Z" }, "id": "CAx11gMjzzjs" }, "outputs": [], "source": [ "@tf.function\n", "def train_step(model, x, y, optimizer, metrics):\n", " with tf.GradientTape() as tape:\n", " logits = model(x, training=True)\n", " # tf.reduce_sum sums the batch sharded per-example loss to a replicated\n", " # global loss (scalar).\n", " loss = tf.reduce_sum(tf.keras.losses.sparse_categorical_crossentropy(\n", " y, logits, from_logits=True))\n", " \n", " gradients = tape.gradient(loss, model.trainable_variables)\n", " optimizer.apply_gradients(zip(gradients, model.trainable_variables))\n", "\n", " for metric in metrics.values():\n", " metric.update_state(y_true=y, y_pred=logits)\n", "\n", " loss_per_sample = loss / len(x)\n", " results = {'loss': loss_per_sample}\n", " return results" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.691615Z", "iopub.status.busy": "2024-01-11T18:16:57.691352Z", "iopub.status.idle": "2024-01-11T18:16:57.695747Z", "shell.execute_reply": "2024-01-11T18:16:57.695146Z" }, "id": "maSTWeRemO0P" }, "outputs": [], "source": [ "@tf.function\n", "def eval_step(model, x, y, metrics):\n", " logits = model(x, training=False)\n", " loss = tf.reduce_sum(tf.keras.losses.sparse_categorical_crossentropy(\n", " y, logits, from_logits=True))\n", "\n", " for metric in metrics.values():\n", " metric.update_state(y_true=y, y_pred=logits)\n", "\n", " loss_per_sample = loss / len(x)\n", " results = {'eval_loss': loss_per_sample}\n", " return results" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.698789Z", "iopub.status.busy": "2024-01-11T18:16:57.698566Z", "iopub.status.idle": "2024-01-11T18:16:57.702339Z", "shell.execute_reply": "2024-01-11T18:16:57.701696Z" }, "id": "dt00axcLmvLr" }, "outputs": [], "source": [ "def pack_dtensor_inputs(images, labels, image_layout, label_layout):\n", " num_local_devices = image_layout.mesh.num_local_devices()\n", " images = tf.split(images, num_local_devices)\n", " labels = tf.split(labels, num_local_devices)\n", " images = dtensor.pack(images, image_layout)\n", " labels = dtensor.pack(labels, label_layout)\n", " return images, labels" ] }, { "cell_type": "markdown", "metadata": { "id": "9Eb-qIJGrxB9" }, "source": [ "## Metric と Optimizer\n", "\n", "Keras `Metric` と `Optimizer` を使って DTensor API を使用する場合、追加のメッシュ情報を指定して、内部状態変数とテンソルがモデルの変数と連携できるようにする必要があります。\n", "\n", "- オプティマイザの場合、DTensor は `keras.dtensor.experimental.optimizers` という新しい実験的な名前空間を使用します。多くの既存の Keras Optimizer は、追加の `mesh` 引数を受け取るように拡張されます。今後のリリースでは、Keras のコアオプティマイザにマージされる可能性があります。\n", "\n", "- 指標の場合、DTensor 対応の `Metric` になるように、コンストラクタに直接引数として `mesh` を指定できます。" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.705526Z", "iopub.status.busy": "2024-01-11T18:16:57.705300Z", "iopub.status.idle": "2024-01-11T18:16:57.770967Z", "shell.execute_reply": "2024-01-11T18:16:57.770266Z" }, "id": "1lu_0mz1sxrl" }, "outputs": [], "source": [ "optimizer = tf.keras.dtensor.experimental.optimizers.Adam(0.01, mesh=mesh)\n", "metrics = {'accuracy': tf.keras.metrics.SparseCategoricalAccuracy(mesh=mesh)}\n", "eval_metrics = {'eval_accuracy': tf.keras.metrics.SparseCategoricalAccuracy(mesh=mesh)}" ] }, { "cell_type": "markdown", "metadata": { "id": "QzufrkistELx" }, "source": [ "## モデルのトレーニング\n", "\n", "以下の例では、batch 次元で入力パイプラインのデータをシャード化し、完全に複製された重みをもつモデルを使ってトレーニングします。\n", "\n", "モデルは 3 つのエポックで、約 97% の精度を達成します。" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:16:57.774613Z", "iopub.status.busy": "2024-01-11T18:16:57.774368Z", "iopub.status.idle": "2024-01-11T18:17:14.400603Z", "shell.execute_reply": "2024-01-11T18:17:14.399858Z" }, "id": "kZW568Dk0vvL" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "============================\n", "Epoch: 0\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\r", " 0/Unknown - 3s 0s/step - loss: 2.4019 - accuracy: 0.0859" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 4/Unknown - 3s 692ms/step - loss: 2.1233 - accuracy: 0.1611" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 10/Unknown - 3s 282ms/step - loss: 1.7960 - accuracy: 0.2829" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 180ms/step - loss: 1.5757 - accuracy: 0.3551" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 22/Unknown - 3s 133ms/step - loss: 1.3916 - accuracy: 0.4056" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 106ms/step - loss: 1.2480 - accuracy: 0.4456" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 89ms/step - loss: 1.1229 - accuracy: 0.4786 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 77ms/step - loss: 1.0421 - accuracy: 0.5064" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 68ms/step - loss: 0.9624 - 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\r", " 52/Unknown - 3s 61ms/step - loss: 0.8969 - 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", " 58/Unknown - 3s 56ms/step - loss: 0.8472 - accuracy: 0.5694" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 51ms/step - loss: 0.8005 - accuracy: 0.5858" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 48ms/step - loss: 0.7649 - accuracy: 0.6006" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 45ms/step - loss: 0.7302 - accuracy: 0.6138" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 82/Unknown - 3s 42ms/step - loss: 0.7035 - accuracy: 0.6257" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 40ms/step - loss: 0.6772 - accuracy: 0.6367" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 38ms/step - loss: 0.6539 - accuracy: 0.6467" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 36ms/step - loss: 0.6312 - accuracy: 0.6560" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 34ms/step - loss: 0.6123 - accuracy: 0.6646" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 33ms/step - loss: 0.5953 - accuracy: 0.6726" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 118/Unknown - 4s 32ms/step - loss: 0.5797 - accuracy: 0.6800" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 31ms/step - loss: 0.5635 - accuracy: 0.6870" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 130/Unknown - 4s 30ms/step - loss: 0.5484 - accuracy: 0.6935" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 29ms/step - loss: 0.5370 - accuracy: 0.6997" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 142/Unknown - 4s 28ms/step - loss: 0.5241 - accuracy: 0.7055" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 148/Unknown - 4s 27ms/step - loss: 0.5134 - accuracy: 0.7109" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 26ms/step - loss: 0.5033 - accuracy: 0.7161" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 26ms/step - loss: 0.4913 - accuracy: 0.7210" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 25ms/step - loss: 0.4830 - accuracy: 0.7257" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 24ms/step - loss: 0.4763 - accuracy: 0.7301" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 178/Unknown - 4s 24ms/step - loss: 0.4685 - accuracy: 0.7343" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 23ms/step - loss: 0.4624 - accuracy: 0.7383" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 23ms/step - loss: 0.4552 - accuracy: 0.7421" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 23ms/step - loss: 0.4493 - accuracy: 0.7457" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 4s 22ms/step - loss: 0.4415 - accuracy: 0.7492" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 208/Unknown - 5s 22ms/step - loss: 0.4361 - accuracy: 0.7526" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 214/Unknown - 5s 21ms/step - loss: 0.4301 - accuracy: 0.7558" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 220/Unknown - 5s 21ms/step - loss: 0.4235 - accuracy: 0.7588" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 5s 21ms/step - loss: 0.4180 - accuracy: 0.7618" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 232/Unknown - 5s 20ms/step - loss: 0.4128 - accuracy: 0.7647" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 5s 20ms/step - loss: 0.4076 - accuracy: 0.7674" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 244/Unknown - 5s 20ms/step - loss: 0.4018 - accuracy: 0.7701" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 5s 20ms/step - loss: 0.3975 - accuracy: 0.7727" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 5s 19ms/step - loss: 0.3934 - accuracy: 0.7752" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 262/Unknown - 5s 19ms/step - loss: 0.3883 - accuracy: 0.7776" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 268/Unknown - 5s 19ms/step - loss: 0.3839 - 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\b\r", " 274/Unknown - 5s 19ms/step - loss: 0.3798 - accuracy: 0.7821" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 280/Unknown - 5s 18ms/step - loss: 0.3767 - accuracy: 0.7843" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 5s 18ms/step - loss: 0.3729 - accuracy: 0.7864" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 292/Unknown - 5s 18ms/step - loss: 0.3685 - accuracy: 0.7885" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 5s 18ms/step - loss: 0.3649 - accuracy: 0.7905" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 304/Unknown - 5s 18ms/step - loss: 0.3615 - accuracy: 0.7924" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 310/Unknown - 5s 17ms/step - loss: 0.3579 - accuracy: 0.7943" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 316/Unknown - 5s 17ms/step - loss: 0.3540 - accuracy: 0.7962" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 322/Unknown - 5s 17ms/step - loss: 0.3510 - accuracy: 0.7980" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 328/Unknown - 6s 17ms/step - loss: 0.3475 - accuracy: 0.7997" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 334/Unknown - 6s 17ms/step - loss: 0.3434 - accuracy: 0.8014" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 6s 17ms/step - loss: 0.3413 - accuracy: 0.8030" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 346/Unknown - 6s 16ms/step - loss: 0.3380 - 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\r", " 352/Unknown - 6s 16ms/step - loss: 0.3345 - accuracy: 0.8062" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 358/Unknown - 6s 16ms/step - loss: 0.3314 - accuracy: 0.8078" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 364/Unknown - 6s 16ms/step - loss: 0.3281 - 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\r", " 370/Unknown - 6s 16ms/step - loss: 0.3259 - accuracy: 0.8107" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 376/Unknown - 6s 16ms/step - loss: 0.3227 - accuracy: 0.8121" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 382/Unknown - 6s 16ms/step - loss: 0.3194 - accuracy: 0.8135" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 388/Unknown - 6s 16ms/step - loss: 0.3165 - 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\r", " 394/Unknown - 6s 16ms/step - loss: 0.3149 - accuracy: 0.8163" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 400/Unknown - 6s 15ms/step - loss: 0.3125 - 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\r", " 406/Unknown - 6s 15ms/step - loss: 0.3103 - 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\r", " 412/Unknown - 6s 15ms/step - loss: 0.3082 - accuracy: 0.8201" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 418/Unknown - 6s 15ms/step - loss: 0.3063 - accuracy: 0.8213" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 424/Unknown - 6s 15ms/step - loss: 0.3041 - accuracy: 0.8225" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 430/Unknown - 6s 15ms/step - loss: 0.3020 - accuracy: 0.8237" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 436/Unknown - 6s 15ms/step - loss: 0.2998 - accuracy: 0.8249" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 442/Unknown - 7s 15ms/step - loss: 0.2978 - accuracy: 0.8260" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 448/Unknown - 7s 15ms/step - loss: 0.2964 - accuracy: 0.8271" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 7s 15ms/step - loss: 0.2948 - accuracy: 0.8282" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 460/Unknown - 7s 15ms/step - loss: 0.2935 - 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\r", " 466/Unknown - 7s 14ms/step - loss: 0.2917 - accuracy: 0.8303" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 468/Unknown - 7s 15ms/step - loss: 0.2910 - accuracy: 0.8306" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 469/Unknown - 7s 15ms/step - loss: 0.2907 - accuracy: 0.8308\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "loss: 0.12670570611953735\n", "accuracy: 0.9109166860580444\n", "eval_loss: 0.04795415699481964\n", "eval_accuracy: 0.958899974822998\n", "============================\n", "Epoch: 1\n", "\r", " 0/Unknown - 0s 0s/step - loss: 0.1073 - accuracy: 0.9766" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 6/Unknown - 0s 23ms/step - loss: 0.1262 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 16ms/step - loss: 0.1237 - accuracy: 0.9649" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 13ms/step - loss: 0.1337 - accuracy: 0.9641" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 0s 12ms/step - loss: 0.1317 - accuracy: 0.9632" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 30/Unknown - 0s 12ms/step - loss: 0.1264 - accuracy: 0.9629" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 11ms/step - loss: 0.1273 - accuracy: 0.9625" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 42/Unknown - 0s 11ms/step - loss: 0.1245 - accuracy: 0.9622" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 1s 10ms/step - loss: 0.1233 - accuracy: 0.9621" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 10ms/step - loss: 0.1263 - accuracy: 0.9620" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 60/Unknown - 1s 10ms/step - loss: 0.1256 - accuracy: 0.9620" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 66/Unknown - 1s 10ms/step - loss: 0.1249 - accuracy: 0.9619" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 10ms/step - loss: 0.1283 - accuracy: 0.9618" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 1s 10ms/step - loss: 0.1276 - accuracy: 0.9617" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 84/Unknown - 1s 10ms/step - loss: 0.1252 - accuracy: 0.9616" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 90/Unknown - 1s 10ms/step - loss: 0.1265 - accuracy: 0.9616" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 10ms/step - loss: 0.1287 - accuracy: 0.9615" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1305 - accuracy: 0.9613 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 108/Unknown - 1s 9ms/step - loss: 0.1322 - accuracy: 0.9612" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1335 - accuracy: 0.9610" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1356 - accuracy: 0.9608" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 126/Unknown - 1s 9ms/step - loss: 0.1381 - accuracy: 0.9606" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1372 - accuracy: 0.9604" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 138/Unknown - 1s 9ms/step - loss: 0.1383 - accuracy: 0.9603" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 1s 9ms/step - loss: 0.1371 - accuracy: 0.9601" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1368 - accuracy: 0.9600" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1359 - accuracy: 0.9599" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 1s 9ms/step - loss: 0.1354 - accuracy: 0.9599" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1351 - accuracy: 0.9598" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1339 - accuracy: 0.9597" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 180/Unknown - 2s 9ms/step - loss: 0.1341 - accuracy: 0.9597" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 186/Unknown - 2s 9ms/step - loss: 0.1347 - accuracy: 0.9597" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1349 - accuracy: 0.9596" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 198/Unknown - 2s 9ms/step - loss: 0.1344 - accuracy: 0.9596" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 204/Unknown - 2s 9ms/step - loss: 0.1349 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 210/Unknown - 2s 9ms/step - loss: 0.1346 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 216/Unknown - 2s 9ms/step - loss: 0.1349 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1341 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 228/Unknown - 2s 9ms/step - loss: 0.1345 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1346 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1355 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 246/Unknown - 2s 9ms/step - loss: 0.1355 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 252/Unknown - 2s 9ms/step - loss: 0.1358 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 258/Unknown - 2s 9ms/step - loss: 0.1353 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1350 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 270/Unknown - 2s 9ms/step - loss: 0.1340 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 2s 9ms/step - loss: 0.1344 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 282/Unknown - 3s 9ms/step - loss: 0.1340 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 288/Unknown - 3s 9ms/step - loss: 0.1338 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 294/Unknown - 3s 9ms/step - loss: 0.1332 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 300/Unknown - 3s 9ms/step - loss: 0.1330 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 306/Unknown - 3s 9ms/step - loss: 0.1327 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.1325 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 318/Unknown - 3s 9ms/step - loss: 0.1324 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 324/Unknown - 3s 9ms/step - loss: 0.1321 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 330/Unknown - 3s 9ms/step - loss: 0.1318 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.1314 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 342/Unknown - 3s 9ms/step - loss: 0.1311 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 348/Unknown - 3s 9ms/step - loss: 0.1308 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 354/Unknown - 3s 9ms/step - loss: 0.1302 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 360/Unknown - 3s 9ms/step - loss: 0.1296 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 366/Unknown - 3s 9ms/step - loss: 0.1291 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 372/Unknown - 3s 9ms/step - loss: 0.1290 - accuracy: 0.9592" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.1283 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 384/Unknown - 3s 9ms/step - loss: 0.1277 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 390/Unknown - 3s 9ms/step - loss: 0.1268 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 396/Unknown - 4s 9ms/step - loss: 0.1266 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 402/Unknown - 4s 9ms/step - loss: 0.1263 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 408/Unknown - 4s 9ms/step - loss: 0.1264 - accuracy: 0.9593" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 414/Unknown - 4s 9ms/step - loss: 0.1262 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 420/Unknown - 4s 9ms/step - loss: 0.1261 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 426/Unknown - 4s 9ms/step - loss: 0.1262 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 432/Unknown - 4s 9ms/step - loss: 0.1262 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 438/Unknown - 4s 9ms/step - loss: 0.1270 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 4s 9ms/step - loss: 0.1274 - accuracy: 0.9594" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 450/Unknown - 4s 9ms/step - loss: 0.1280 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 456/Unknown - 4s 9ms/step - loss: 0.1278 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 462/Unknown - 4s 9ms/step - loss: 0.1288 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 468/Unknown - 4s 9ms/step - loss: 0.1286 - accuracy: 0.9595" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 469/Unknown - 4s 9ms/step - loss: 0.1285 - accuracy: 0.9595\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "loss: 0.08036476373672485\n", "accuracy: 0.9599999785423279\n", "eval_loss: 0.008920179679989815\n", "eval_accuracy: 0.9642000198364258\n", "============================\n", "Epoch: 2\n", "\r", " 0/Unknown - 0s 0s/step - loss: 0.1102 - accuracy: 0.9688" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 6/Unknown - 0s 23ms/step - loss: 0.1375 - accuracy: 0.9626" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 16ms/step - loss: 0.1137 - accuracy: 0.9621" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 13ms/step - loss: 0.1053 - accuracy: 0.9633" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 0s 12ms/step - loss: 0.1093 - accuracy: 0.9640" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 30/Unknown - 0s 11ms/step - loss: 0.1113 - accuracy: 0.9643" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 0s 11ms/step - loss: 0.1071 - accuracy: 0.9647" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 42/Unknown - 0s 11ms/step - loss: 0.1050 - accuracy: 0.9651" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 0s 10ms/step - loss: 0.1055 - accuracy: 0.9654" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 10ms/step - loss: 0.1075 - accuracy: 0.9655" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 60/Unknown - 1s 10ms/step - loss: 0.1094 - accuracy: 0.9656" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 66/Unknown - 1s 10ms/step - loss: 0.1070 - accuracy: 0.9658" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 10ms/step - loss: 0.1059 - accuracy: 0.9659" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 1s 10ms/step - loss: 0.1054 - accuracy: 0.9661" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 84/Unknown - 1s 10ms/step - loss: 0.1041 - accuracy: 0.9662" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 90/Unknown - 1s 10ms/step - loss: 0.1039 - accuracy: 0.9663" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1029 - accuracy: 0.9664 " ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1004 - accuracy: 0.9665" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 108/Unknown - 1s 9ms/step - loss: 0.1003 - accuracy: 0.9667" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1011 - accuracy: 0.9668" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.1007 - accuracy: 0.9669" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 126/Unknown - 1s 9ms/step - loss: 0.0996 - accuracy: 0.9670" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.0982 - accuracy: 0.9671" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 138/Unknown - 1s 9ms/step - loss: 0.0997 - accuracy: 0.9672" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.0996 - accuracy: 0.9672" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.0985 - accuracy: 0.9673" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 157/Unknown - 1s 9ms/step - loss: 0.0987 - accuracy: 0.9674" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 1s 9ms/step - loss: 0.0990 - accuracy: 0.9674" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.1003 - accuracy: 0.9675" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 175/Unknown - 2s 9ms/step - loss: 0.0997 - accuracy: 0.9675" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 2s 9ms/step - loss: 0.0989 - accuracy: 0.9675" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0997 - accuracy: 0.9676" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0995 - accuracy: 0.9676" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0991 - accuracy: 0.9676" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 205/Unknown - 2s 9ms/step - loss: 0.0986 - accuracy: 0.9677" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0992 - accuracy: 0.9677" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 217/Unknown - 2s 9ms/step - loss: 0.1001 - accuracy: 0.9677" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 223/Unknown - 2s 9ms/step - loss: 0.0990 - accuracy: 0.9678" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 229/Unknown - 2s 9ms/step - loss: 0.0990 - accuracy: 0.9678" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 235/Unknown - 2s 9ms/step - loss: 0.0981 - accuracy: 0.9678" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0988 - accuracy: 0.9679" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0984 - accuracy: 0.9679" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 253/Unknown - 2s 9ms/step - loss: 0.0990 - accuracy: 0.9680" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 259/Unknown - 2s 9ms/step - loss: 0.0989 - accuracy: 0.9680" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 265/Unknown - 2s 9ms/step - loss: 0.0994 - accuracy: 0.9680" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 2s 9ms/step - loss: 0.0987 - accuracy: 0.9680" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 277/Unknown - 2s 9ms/step - loss: 0.0985 - accuracy: 0.9681" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.0979 - accuracy: 0.9681" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 289/Unknown - 3s 9ms/step - loss: 0.0984 - accuracy: 0.9681" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 3s 9ms/step - loss: 0.0984 - accuracy: 0.9681" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.0979 - accuracy: 0.9681" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 307/Unknown - 3s 9ms/step - loss: 0.0982 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.0979 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 319/Unknown - 3s 9ms/step - loss: 0.0980 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 325/Unknown - 3s 9ms/step - loss: 0.0984 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 331/Unknown - 3s 9ms/step - loss: 0.0989 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 337/Unknown - 3s 9ms/step - loss: 0.0990 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 343/Unknown - 3s 9ms/step - loss: 0.0992 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 349/Unknown - 3s 9ms/step - loss: 0.0994 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.0998 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 361/Unknown - 3s 9ms/step - loss: 0.0998 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 367/Unknown - 3s 9ms/step - loss: 0.0996 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 373/Unknown - 3s 9ms/step - loss: 0.0996 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 379/Unknown - 3s 9ms/step - loss: 0.1000 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 385/Unknown - 3s 9ms/step - loss: 0.1007 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 391/Unknown - 3s 9ms/step - loss: 0.1011 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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/Unknown - 3s 9ms/step - loss: 0.1006 - accuracy: 0.9683" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 403/Unknown - 4s 9ms/step - loss: 0.1005 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 409/Unknown - 4s 9ms/step - loss: 0.1011 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 415/Unknown - 4s 9ms/step - loss: 0.1013 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 421/Unknown - 4s 9ms/step - loss: 0.1016 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 427/Unknown - 4s 9ms/step - loss: 0.1020 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 433/Unknown - 4s 9ms/step - loss: 0.1021 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 439/Unknown - 4s 9ms/step - loss: 0.1019 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 445/Unknown - 4s 9ms/step - loss: 0.1017 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 451/Unknown - 4s 9ms/step - loss: 0.1014 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 457/Unknown - 4s 9ms/step - loss: 0.1016 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\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 - 4s 9ms/step - loss: 0.1015 - accuracy: 0.9682" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r", " 469/Unknown - 4s 9ms/step - loss: 0.1010 - accuracy: 0.9682\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "loss: 0.044021397829055786\n", "accuracy: 0.9682833552360535\n", "eval_loss: 0.05413995310664177\n", "eval_accuracy: 0.9656000137329102\n" ] } ], "source": [ "num_epochs = 3\n", "\n", "image_layout = dtensor.Layout.batch_sharded(mesh, 'batch', rank=4)\n", "label_layout = dtensor.Layout.batch_sharded(mesh, 'batch', rank=1)\n", "\n", "for epoch in range(num_epochs):\n", " print(\"============================\") \n", " print(\"Epoch: \", epoch)\n", " for metric in metrics.values():\n", " metric.reset_state()\n", " step = 0\n", " results = {}\n", " pbar = tf.keras.utils.Progbar(target=None, stateful_metrics=[])\n", " for input in ds_train:\n", " images, labels = input[0], input[1]\n", " images, labels = pack_dtensor_inputs(\n", " images, labels, image_layout, label_layout)\n", "\n", " results.update(train_step(model, images, labels, optimizer, metrics))\n", " for metric_name, metric in metrics.items():\n", " results[metric_name] = metric.result()\n", "\n", " pbar.update(step, values=results.items(), finalize=False)\n", " step += 1\n", " pbar.update(step, values=results.items(), finalize=True)\n", "\n", " for metric in eval_metrics.values():\n", " metric.reset_state()\n", " for input in ds_test:\n", " images, labels = input[0], input[1]\n", " images, labels = pack_dtensor_inputs(\n", " images, labels, image_layout, label_layout)\n", " results.update(eval_step(model, images, labels, eval_metrics))\n", "\n", " for metric_name, metric in eval_metrics.items():\n", " results[metric_name] = metric.result()\n", " \n", " for metric_name, metric in results.items():\n", " print(f\"{metric_name}: {metric.numpy()}\")\n" ] }, { "cell_type": "markdown", "metadata": { "id": "HYEXF6qCuoSr" }, "source": [ "## 既存のモデルコードのレイアウトを指定する\n", "\n", "ほとんどの場合、モデルは特定のユースケースでうまく動作するようになっているため、モデル内の個別のレイヤーに `Layout` 情報を指定する作業は膨大であり、多数の編集作業が必要となります。\n", "\n", "既存の Keras モデルを DTensor API で動作できるようにするための変換作業を行いやすくするために、グローバルな観点で `Layout` を指定できる新しい `dtensor.LayoutMap` API を使用できます。\n", "\n", "まず、`LayoutMap` インスタンスを作成する必要があります。これは、モデルの重みに指定するすべての `Layout` を構成するディクショナリのようなオブジェクトです。\n", "\n", "`LayoutMap` には、init 時に `Mesh` インスタンスが必要です。これは、Layout が構成されていない、任意の重みに対するデフォルトの複製済み `Layout` を指定するために使用できます。すべてのモデルの重みを完全に複製するだけの場合は、空の `LayoutMap` を指定すると、デフォルトのメッシュを使って複製された `Layout` が作成されます。\n", "\n", "`LayoutMap` は、文字列をキーとして、`Layout` を値として使用します。通常の Python dict とこのクラスでは、動作が異なります。文字列キーは、値を取得する際の正規表現として処理されます。" ] }, { "cell_type": "markdown", "metadata": { "id": "SCq5Nl-UP_dS" }, "source": [ "### Subclassed モデル\n", "\n", "Keras のサブクラス化モデル構文を使って定義された以下のモデルについて考察してみましょう。" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.404581Z", "iopub.status.busy": "2024-01-11T18:17:14.404021Z", "iopub.status.idle": "2024-01-11T18:17:14.409049Z", "shell.execute_reply": "2024-01-11T18:17:14.408484Z" }, "id": "LZ0hRFs8unu0" }, "outputs": [], "source": [ "class SubclassedModel(tf.keras.Model):\n", "\n", " def __init__(self, name=None):\n", " super().__init__(name=name)\n", " self.feature = tf.keras.layers.Dense(16)\n", " self.feature_2 = tf.keras.layers.Dense(24)\n", " self.dropout = tf.keras.layers.Dropout(0.1)\n", "\n", " def call(self, inputs, training=None):\n", " x = self.feature(inputs)\n", " x = self.dropout(x, training=training)\n", " return self.feature_2(x)" ] }, { "cell_type": "markdown", "metadata": { "id": "1njxqPB-yS97" }, "source": [ "このモデルには、2 つの `Dense` レイヤーに対し `kernel` と `bias` レイヤーという 4 つの重みがあります。それぞれは、オブジェクトパスに基づいてマッピングされています。\n", "\n", "- `model.feature.kernel`\n", "- `model.feature.bias`\n", "- `model.feature_2.kernel`\n", "- `model.feature_2.bias`\n", "\n", "注意: Subclassed モデルでは、マッピングから Layout を取得する際に、レイヤーの `.name` 属性ではなく、属性名がキーとして使用されます。これは、`tf.Module` のチェックポイント設定が使う規則と同じです。多数のレイヤーを持つ複雑なモデルでは、[チェックポイントを手動で検査する](https://www.tensorflow.org/guide/checkpoint#manually_inspecting_checkpoints)ことで、属性のマッピングを確認できます。\n", "\n", "では、以下の `LayoutMap` を定義して、モデルを適用しましょう。" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.412548Z", "iopub.status.busy": "2024-01-11T18:17:14.411936Z", "iopub.status.idle": "2024-01-11T18:17:14.433807Z", "shell.execute_reply": "2024-01-11T18:17:14.433121Z" }, "id": "goVX6iIZw468" }, "outputs": [], "source": [ "layout_map = tf.keras.dtensor.experimental.LayoutMap(mesh=mesh)\n", "\n", "layout_map['feature.*kernel'] = dtensor.Layout.batch_sharded(mesh, 'batch', rank=2)\n", "layout_map['feature.*bias'] = dtensor.Layout.batch_sharded(mesh, 'batch', rank=1)\n", "\n", "with layout_map.scope():\n", " subclassed_model = SubclassedModel()" ] }, { "cell_type": "markdown", "metadata": { "id": "M32HcSp_PyWs" }, "source": [ "モデルの重みは最初のセルに作成されているため、DTensor 入力でモデルを呼び出し、重みに期待されるレイアウトがあることを確認します。" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.437255Z", "iopub.status.busy": "2024-01-11T18:17:14.436722Z", "iopub.status.idle": "2024-01-11T18:17:14.709766Z", "shell.execute_reply": "2024-01-11T18:17:14.708949Z" }, "id": "c3CbD9l7qUNq" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Layout.from_string(sharding_specs:batch,unsharded, mesh:|batch=8|0,1,2,3,4,5,6,7|0,1,2,3,4,5,6,7|/job:localhost/replica:0/task:0/device:CPU:0,/job:localhost/replica:0/task:0/device:CPU:1,/job:localhost/replica:0/task:0/device:CPU:2,/job:localhost/replica:0/task:0/device:CPU:3,/job:localhost/replica:0/task:0/device:CPU:4,/job:localhost/replica:0/task:0/device:CPU:5,/job:localhost/replica:0/task:0/device:CPU:6,/job:localhost/replica:0/task:0/device:CPU:7)\n" ] } ], "source": [ "dtensor_input = dtensor.copy_to_mesh(tf.zeros((16, 16)), layout=unsharded_layout_2d)\n", "# Trigger the weights creation for subclass model\n", "subclassed_model(dtensor_input)\n", "\n", "print(subclassed_model.feature.kernel.layout)" ] }, { "cell_type": "markdown", "metadata": { "id": "ZyCnfd-4Q2jk" }, "source": [ "これにより、既存のコードを更新することなく、`Layout` をすばやくモデルにマッピングすることができます。 " ] }, { "cell_type": "markdown", "metadata": { "id": "6GliUdWTQnKC" }, "source": [ "### Sequential モデルと Functional モデル" ] }, { "cell_type": "markdown", "metadata": { "id": "6zzvTqAR2Teu" }, "source": [ "Keras Functional モデルと Sequential モデルの場合も、`LayoutMap` を使用できます。\n", "\n", "注意: Functional モデルと Sequential モデルでは、マッピングにわずかな違いがあります。モデルのレイヤーには、モデルに接続された公開属性がありません（ただし、`model.layers` を介してリストとしてアクセス可能です）。この場合、文字列名をキーとして使用します。文字列名は、モデル内で必ず一意の値です。" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.713473Z", "iopub.status.busy": "2024-01-11T18:17:14.713196Z", "iopub.status.idle": "2024-01-11T18:17:14.717255Z", "shell.execute_reply": "2024-01-11T18:17:14.716679Z" }, "id": "gXK2EquIRJCC" }, "outputs": [], "source": [ "layout_map = tf.keras.dtensor.experimental.LayoutMap(mesh=mesh)\n", "\n", "layout_map['feature.*kernel'] = dtensor.Layout.batch_sharded(mesh, 'batch', rank=2)\n", "layout_map['feature.*bias'] = dtensor.Layout.batch_sharded(mesh, 'batch', rank=1)" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.720034Z", "iopub.status.busy": "2024-01-11T18:17:14.719643Z", "iopub.status.idle": "2024-01-11T18:17:14.839204Z", "shell.execute_reply": "2024-01-11T18:17:14.838546Z" }, "id": "cBzwJqrg2TH3" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Layout.from_string(sharding_specs:batch,unsharded, mesh:|batch=8|0,1,2,3,4,5,6,7|0,1,2,3,4,5,6,7|/job:localhost/replica:0/task:0/device:CPU:0,/job:localhost/replica:0/task:0/device:CPU:1,/job:localhost/replica:0/task:0/device:CPU:2,/job:localhost/replica:0/task:0/device:CPU:3,/job:localhost/replica:0/task:0/device:CPU:4,/job:localhost/replica:0/task:0/device:CPU:5,/job:localhost/replica:0/task:0/device:CPU:6,/job:localhost/replica:0/task:0/device:CPU:7)\n" ] } ], "source": [ "with layout_map.scope():\n", " inputs = tf.keras.Input((16,), batch_size=16)\n", " x = tf.keras.layers.Dense(16, name='feature')(inputs)\n", " x = tf.keras.layers.Dropout(0.1)(x)\n", " output = tf.keras.layers.Dense(32, name='feature_2')(x)\n", " model = tf.keras.Model(inputs, output)\n", "\n", "print(model.layers[1].kernel.layout)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "execution": { "iopub.execute_input": "2024-01-11T18:17:14.842584Z", "iopub.status.busy": "2024-01-11T18:17:14.841977Z", "iopub.status.idle": "2024-01-11T18:17:14.909300Z", "shell.execute_reply": "2024-01-11T18:17:14.908590Z" }, "id": "pPuh1NlE3-wO" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Layout.from_string(sharding_specs:batch,unsharded, mesh:|batch=8|0,1,2,3,4,5,6,7|0,1,2,3,4,5,6,7|/job:localhost/replica:0/task:0/device:CPU:0,/job:localhost/replica:0/task:0/device:CPU:1,/job:localhost/replica:0/task:0/device:CPU:2,/job:localhost/replica:0/task:0/device:CPU:3,/job:localhost/replica:0/task:0/device:CPU:4,/job:localhost/replica:0/task:0/device:CPU:5,/job:localhost/replica:0/task:0/device:CPU:6,/job:localhost/replica:0/task:0/device:CPU:7)\n" ] } ], "source": [ "with layout_map.scope():\n", " model = tf.keras.Sequential([\n", " tf.keras.layers.Dense(16, name='feature', input_shape=(16,)),\n", " tf.keras.layers.Dropout(0.1),\n", " tf.keras.layers.Dense(32, name='feature_2')\n", " ])\n", "\n", "print(model.layers[2].kernel.layout)" ] } ], "metadata": { "colab": { "name": "dtensor_keras_tutorial.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": "3.9.18" } }, "nbformat": 4, "nbformat_minor": 0 }