{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "colab": { "provenance": [], "gpuType": "T4" }, "kernelspec": { "name": "python3", "display_name": "Python 3" }, "language_info": { "name": "python" }, "accelerator": "GPU" }, "cells": [ { "cell_type": "code", "source": [ "# Import necessary libraries\n", "import numpy as np\n", "import tensorflow as tf\n", "from tensorflow import keras\n", "from tensorflow.keras import layers\n", "import matplotlib.pyplot as plt\n", "\n", "# Load CIFAR-10 dataset\n", "(x_train, _), (x_test, _) = keras.datasets.cifar10.load_data()\n", "x_train = x_train.astype(\"float32\") / 255.0\n", "x_test = x_test.astype(\"float32\") / 255.0\n", "img_shape = x_train.shape[1:]\n", "\n", "latent_dim = 64\n", "\n", "# Define Encoder\n", "encoder_inputs = keras.Input(shape=img_shape)\n", "x = layers.Conv2D(32, 3, activation=\"relu\", strides=2, padding=\"same\")(encoder_inputs)\n", "x = layers.Conv2D(64, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n", "x = layers.Flatten()(x)\n", "x = layers.Dense(128, activation=\"relu\")(x)\n", "z_mean = layers.Dense(latent_dim)(x)\n", "z_log_var = layers.Dense(latent_dim)(x)\n", "\n", "# Sampling Layer\n", "class Sampling(layers.Layer):\n", " def call(self, inputs):\n", " z_mean, z_log_var = inputs\n", " epsilon = tf.keras.backend.random_normal(shape=(tf.shape(z_mean)[0], latent_dim))\n", " return z_mean + tf.exp(0.5 * z_log_var) * epsilon\n", "\n", "z = Sampling()([z_mean, z_log_var])\n", "\n", "# Define Decoder\n", "decoder_inputs = keras.Input(shape=(latent_dim,))\n", "x = layers.Dense(128, activation=\"relu\")(decoder_inputs)\n", "x = layers.Dense(np.prod(img_shape), activation=\"sigmoid\")(x)\n", "x = layers.Reshape(img_shape)(x)\n", "decoder = keras.Model(decoder_inputs, x, name=\"decoder\")\n", "\n", "# Define VAE Model\n", "vae = keras.Model(encoder_inputs, decoder(z), name=\"vae\")\n", "\n", "# Compile VAE\n", "vae.compile(optimizer=keras.optimizers.Adam(), loss=\"mse\")\n", "\n", "# Train VAE\n", "vae.fit(x_train, x_train, epochs=500, batch_size=128, validation_data=(x_test, x_test))\n", "\n", "# Generate Alternative Images\n", "def generate_images(num_samples=10):\n", " random_latent_vectors = np.random.normal(size=(num_samples, latent_dim))\n", " generated_images = decoder.predict(random_latent_vectors)\n", " return generated_images\n", "\n", "# Show Generated Images\n", "dream_images = generate_images()\n", "plt.figure(figsize=(10, 10))\n", "for i in range(9):\n", " plt.subplot(3, 3, i + 1)\n", " plt.imshow(dream_images[i])\n", " plt.axis(\"off\")\n", "plt.show()\n" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "PQyK2VmQkhN7", "outputId": "7d0f5b0a-ebc2-49ac-82bd-fe3036e9a84f" }, "execution_count": 10, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Epoch 1/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 14ms/step - loss: 0.0387 - val_loss: 0.0176\n", "Epoch 2/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 6ms/step - loss: 0.0165 - val_loss: 0.0140\n", "Epoch 3/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0135 - val_loss: 0.0126\n", "Epoch 4/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0123 - val_loss: 0.0118\n", "Epoch 5/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0116 - val_loss: 0.0114\n", "Epoch 6/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0113 - val_loss: 0.0113\n", "Epoch 7/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0110 - val_loss: 0.0108\n", "Epoch 8/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0107 - val_loss: 0.0105\n", "Epoch 9/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0105 - val_loss: 0.0104\n", "Epoch 10/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0102 - val_loss: 0.0102\n", "Epoch 11/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0101 - val_loss: 0.0101\n", "Epoch 12/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0100 - val_loss: 0.0099\n", "Epoch 13/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0099 - val_loss: 0.0099\n", "Epoch 14/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0098 - val_loss: 0.0098\n", "Epoch 15/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0097 - val_loss: 0.0097\n", "Epoch 16/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0096 - val_loss: 0.0096\n", "Epoch 17/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0095 - val_loss: 0.0095\n", "Epoch 18/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0094 - val_loss: 0.0095\n", "Epoch 19/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0094 - val_loss: 0.0095\n", "Epoch 20/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0094 - val_loss: 0.0093\n", "Epoch 21/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0093 - val_loss: 0.0093\n", "Epoch 22/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0092 - val_loss: 0.0092\n", "Epoch 23/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0091 - val_loss: 0.0092\n", "Epoch 24/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0091 - val_loss: 0.0091\n", "Epoch 25/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0091 - val_loss: 0.0090\n", "Epoch 26/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0090 - val_loss: 0.0090\n", "Epoch 27/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0090 - val_loss: 0.0090\n", "Epoch 28/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0089 - val_loss: 0.0090\n", "Epoch 29/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0089 - val_loss: 0.0089\n", "Epoch 30/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0088 - val_loss: 0.0088\n", "Epoch 31/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0088 - val_loss: 0.0088\n", "Epoch 32/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0087 - val_loss: 0.0088\n", "Epoch 33/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0087 - val_loss: 0.0088\n", "Epoch 34/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0087 - val_loss: 0.0087\n", "Epoch 35/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0086 - val_loss: 0.0087\n", "Epoch 36/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0086 - val_loss: 0.0087\n", "Epoch 37/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0086 - val_loss: 0.0087\n", "Epoch 38/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0086 - val_loss: 0.0087\n", "Epoch 39/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0086 - val_loss: 0.0086\n", "Epoch 40/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0085 - val_loss: 0.0086\n", "Epoch 41/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 42/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0086\n", "Epoch 43/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0086\n", "Epoch 44/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 45/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 46/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 47/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 48/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0086\n", "Epoch 49/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 50/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 51/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 52/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0085 - val_loss: 0.0085\n", "Epoch 53/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 54/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 55/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 56/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0084\n", "Epoch 57/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0085\n", "Epoch 58/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0084\n", "Epoch 59/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0084\n", "Epoch 60/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0084 - val_loss: 0.0084\n", "Epoch 61/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 62/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 63/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 64/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 65/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 66/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 67/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 68/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 69/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 70/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 71/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 72/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 73/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 74/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 75/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 76/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 77/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 78/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 79/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 80/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 81/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 82/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 83/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 84/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 85/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 86/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 87/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 88/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 89/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 90/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 91/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0084\n", "Epoch 92/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0084\n", "Epoch 93/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 94/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 95/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 96/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 97/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 98/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 99/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 100/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 101/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 102/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 103/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 104/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0083 - val_loss: 0.0083\n", "Epoch 105/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 106/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 107/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 108/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 109/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 110/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 111/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 112/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 113/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 114/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 115/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 116/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 117/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 118/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 119/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 120/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 121/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 122/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 123/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 124/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 125/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 126/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 127/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 128/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 129/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 130/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 131/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 132/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 133/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 134/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 135/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 136/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 137/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 138/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 139/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 140/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 141/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 142/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 143/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 144/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 145/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 146/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 147/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 148/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 149/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 150/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 151/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 152/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 153/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 154/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 155/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 156/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 157/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 158/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 159/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 160/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 161/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 162/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 163/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 164/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 165/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 166/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0082\n", "Epoch 167/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 168/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 169/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 170/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 171/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 172/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 173/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 174/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 175/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 176/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 177/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 178/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 179/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 180/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 181/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 182/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0082\n", "Epoch 183/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 184/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 185/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 186/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 187/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 188/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 189/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0082\n", "Epoch 190/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 191/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 192/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 193/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 194/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 195/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 196/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 197/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 198/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 199/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 200/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 201/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 202/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 203/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 204/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 205/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 206/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 207/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 208/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 209/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 210/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 211/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 212/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 213/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 214/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 215/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 216/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 217/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 218/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 219/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 220/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 221/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 222/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 223/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 224/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 225/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 226/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 227/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0082\n", "Epoch 228/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 229/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 230/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 231/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 232/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 233/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 234/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 235/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 236/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 237/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 238/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 239/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 240/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 241/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 242/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 243/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 244/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 245/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 246/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 247/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 248/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 249/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0083\n", "Epoch 250/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 251/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 252/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 253/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 254/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 255/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 256/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 257/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 258/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 259/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 260/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 261/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 262/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 263/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 264/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 265/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 266/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 267/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 268/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 269/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 270/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 271/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 272/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 273/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 274/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 275/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 276/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 277/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 278/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 279/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 280/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 281/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 282/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 283/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 284/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 285/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 286/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 287/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 288/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 289/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 290/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 291/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 292/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 293/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 294/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 295/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 296/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 297/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 298/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 299/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 300/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 301/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 302/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 303/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 304/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 305/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 306/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 307/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 308/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 309/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 310/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 311/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 312/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 313/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 314/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 315/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 316/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 317/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 318/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 319/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 320/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 321/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 322/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 323/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 324/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 325/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 326/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 327/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 328/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 329/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 330/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 331/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 332/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 333/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 334/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 335/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 336/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 337/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 338/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 339/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 340/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 341/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 342/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 343/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 344/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 345/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 346/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 347/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 348/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 349/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 350/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 351/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 352/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 353/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 354/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 355/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 356/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 357/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 358/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0082 - val_loss: 0.0082\n", "Epoch 359/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 360/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 361/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 362/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 363/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 364/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 365/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 366/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 367/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 368/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0083\n", "Epoch 369/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 370/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 371/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 372/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 373/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 374/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 375/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 376/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 377/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 378/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 379/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 380/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 381/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 382/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 383/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 384/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 385/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 386/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 387/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 388/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 389/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 390/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 391/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 392/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 393/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 394/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 395/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 396/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 397/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 398/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 399/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 400/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 401/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 402/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 403/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 404/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 405/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 406/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 407/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 408/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 409/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 410/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 411/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 412/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 413/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 414/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 415/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 416/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 417/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 418/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 419/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 420/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 421/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 422/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 423/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 424/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 425/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 426/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 427/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 428/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 429/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 430/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0081 - val_loss: 0.0082\n", "Epoch 431/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 432/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 433/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 434/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 435/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 436/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 437/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 438/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 439/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 440/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 441/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 442/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 443/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 444/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 445/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 446/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 447/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 448/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 449/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 450/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 451/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 452/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 453/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 454/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 455/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 456/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 457/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 458/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 459/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 460/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 461/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 462/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 463/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 464/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 465/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 466/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 467/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 468/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 469/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 470/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 471/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 472/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 473/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 474/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 475/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 476/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 477/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 478/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 479/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 480/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 481/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 482/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 483/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 484/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 485/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 486/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 7ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 487/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 488/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 489/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 490/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 491/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 492/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 493/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 494/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 495/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 496/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 497/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0082\n", "Epoch 498/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 499/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "Epoch 500/500\n", "\u001b[1m391/391\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 5ms/step - loss: 0.0080 - val_loss: 0.0081\n", "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 117ms/step\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": {} } ] }, { "cell_type": "code", "source": [], "metadata": { "id": "j2yz5A7sle29" }, "execution_count": null, "outputs": [] } ] }