{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "9y222HgfM-OA" }, "source": [ "# 머신 러닝 교과서 3판" ] }, { "cell_type": "markdown", "metadata": { "id": "-5SXCZX_M-OD" }, "source": [ "# 17장 - 새로운 데이터 합성을 위한 생성적 적대 신경망 (1/2)" ] }, { "cell_type": "markdown", "metadata": { "id": "hx1MAB0FbouH" }, "source": [ "**아래 링크를 통해 이 노트북을 주피터 노트북 뷰어(nbviewer.jupyter.org)로 보거나 구글 코랩(colab.research.google.com)에서 실행할 수 있습니다.**\n", "\n", "\n", " \n", " \n", "
\n", " 주피터 노트북 뷰어로 보기\n", " \n", " 구글 코랩(Colab)에서 실행하기\n", "
" ] }, { "cell_type": "markdown", "metadata": { "id": "4wzfKSEhbouH" }, "source": [ "### 목차" ] }, { "cell_type": "markdown", "metadata": { "id": "1vc8_zr4bouH" }, "source": [ "- 생성적 적대 신경망 소개\n", " - 오토인코더\n", " - 새로운 데이터 합성을 위한 생성 모델\n", " - GAN으로 새로운 샘플 생성하기\n", " - GAN의 생성자와 판별자 손실 함수 이해하기\n", "- 밑바닥부터 GAN 모델 구현하기\n", " - 구글 코랩에서 GAN 모델 훈련하기\n", " - 생성자와 판별자 신경망 구현하기\n", " - 훈련 데이터셋 정의하기\n", " - GAN 모델 훈련하기" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "id": "2AfJgihdM-OE" }, "outputs": [], "source": [ "from IPython.display import Image" ] }, { "cell_type": "markdown", "metadata": { "id": "41bk3eM5M-OE" }, "source": [ "# 생성적 적대 신경망 소개\n", "\n", "## 오토인코더" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 470 }, "id": "phbH6zGQM-OE", "outputId": "b080b042-952e-49d5-8afa-b8283e991f42" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQ2', width=500)" ] }, { "cell_type": "markdown", "metadata": { "id": "Cuw0DGR2M-OF" }, "source": [ "## 새로운 데이터 합성을 위한 생성 모델" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 438 }, "id": "ziTpZ3G5M-OF", "outputId": "98e6507c-0901-46bb-dd73-62b6ed853ae5" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQH', width=700)" ] }, { "cell_type": "markdown", "metadata": { "id": "U1DsTJvLM-OF" }, "source": [ "## GAN으로 새로운 샘플 생성하기" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 338 }, "id": "2DijKimVM-OF", "outputId": "8837a45f-ef5d-4483-c2df-e2c88e5e133b" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQ7', width=700)" ] }, { "cell_type": "markdown", "metadata": { "id": "iamOayQfM-OF" }, "source": [ "## GAN의 생성자와 판별자 손실 함수 이해하기" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 516 }, "id": "P0SEKNGpM-OG", "outputId": "3da8d17f-92d6-4332-c70e-401538e0f177" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQF', width=800)" ] }, { "cell_type": "markdown", "metadata": { "id": "mgleEEF1M-OG" }, "source": [ "# 밑바닥부터 GAN 모델 구현하기" ] }, { "cell_type": "markdown", "metadata": { "id": "a1UBoKm_M-OG" }, "source": [ "## 구글 코랩에서 GAN 모델 훈련하기" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 507 }, "id": "sBdI0g22M-OG", "outputId": "e53fc76e-b379-4b71-bee4-172dc0983f89" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQb', width=700)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 526 }, "id": "Mf0pXqraM-OH", "outputId": "f6958ae5-4e28-42ae-9310-dbe38dd7bc19" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQN', width=600)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 533 }, "id": "Nd67sNNIM-OH", "outputId": "7dc62ea7-10e9-47fa-e97c-52cd76cac87f" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQA', width=600)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "_khN3hi1xxQ-", "outputId": "2a45beea-b00a-4fd5-9724-61edaec3880b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2.19.0\n", "GPU 여부: True\n", "/device:GPU:0\n" ] } ], "source": [ "import tensorflow as tf\n", "print(tf.__version__)\n", "\n", "print(\"GPU 여부:\", len(tf.config.list_physical_devices('GPU')) > 0)\n", "\n", "if tf.config.list_physical_devices('GPU'):\n", " device_name = tf.test.gpu_device_name()\n", "else:\n", " device_name = 'cpu:0'\n", "\n", "print(device_name)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "ceNFHOHHNVwz", "outputId": "2c4bb216-05a3-4a5b-9d86-2c806a13c498" }, "outputs": [ { "data": { "text/plain": [ "3314432" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tf.config.experimental.get_memory_info('GPU:0')['current']" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "5sNcfuwHPQMs", "outputId": "eff98c1a-a7d5-4adb-8b0e-9bc0cd44a7c6" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Wed Sep 3 03:17:27 2025 \n", "+-----------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 550.54.15 Driver Version: 550.54.15 CUDA Version: 12.4 |\n", "|-----------------------------------------+------------------------+----------------------+\n", "| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n", "| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n", "| | | MIG M. |\n", "|=========================================+========================+======================|\n", "| 0 NVIDIA A100-SXM4-40GB Off | 00000000:00:04.0 Off | 0 |\n", "| N/A 34C P0 51W / 400W | 423MiB / 40960MiB | 2% Default |\n", "| | | Disabled |\n", "+-----------------------------------------+------------------------+----------------------+\n", " \n", "+-----------------------------------------------------------------------------------------+\n", "| Processes: |\n", "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=========================================================================================|\n", "+-----------------------------------------------------------------------------------------+\n" ] } ], "source": [ "!nvidia-smi" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "4_yH3L4wTOLb", "outputId": "7905415b-72c9-4150-8314-8677b9ddc847" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "No LSB modules are available.\n", "Distributor ID:\tUbuntu\n", "Description:\tUbuntu 22.04.4 LTS\n", "Release:\t22.04\n", "Codename:\tjammy\n" ] } ], "source": [ "!lsb_release -a" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "id": "ehoy80rqwNnl" }, "outputs": [], "source": [ "#from google.colab import drive\n", "#drive.mount('/content/drive/')" ] }, { "cell_type": "markdown", "metadata": { "id": "jHHcQfM2M-OI" }, "source": [ "## 생성자와 판별자 신경망 구현하기" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 474 }, "id": "ig_rZtXqM-OI", "outputId": "7a01e90d-7c4d-42a2-e1df-62df3b1d354c" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQp', width=600)" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 360 }, "id": "uc4FQzczM-OJ", "outputId": "2501e9fb-8672-4c6d-874c-7be1d9d50948" }, "outputs": [ { "data": { "text/html": [ "" ], "text/plain": [ "" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Image(url='https://git.io/JLAQh', width=600)" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "id": "FRh8E66HwXRL" }, "outputs": [], "source": [ "import tensorflow as tf\n", "import tensorflow_datasets as tfds\n", "import numpy as np\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "id": "tOL56FmXwoL-" }, "outputs": [], "source": [ "## 생성자 함수를 정의합니다:\n", "def make_generator_network(\n", " num_hidden_layers=1,\n", " num_hidden_units=100,\n", " num_output_units=784):\n", " model = tf.keras.Sequential()\n", " for i in range(num_hidden_layers):\n", " model.add(\n", " tf.keras.layers.Dense(\n", " units=num_hidden_units,\n", " use_bias=False)\n", " )\n", " model.add(tf.keras.layers.LeakyReLU())\n", "\n", " model.add(tf.keras.layers.Dense(\n", " units=num_output_units, activation='tanh'))\n", " return model\n", "\n", "## 판별자 함수를 정의합니다:\n", "def make_discriminator_network(\n", " num_hidden_layers=1,\n", " num_hidden_units=100,\n", " num_output_units=1):\n", " model = tf.keras.Sequential()\n", " for i in range(num_hidden_layers):\n", " model.add(tf.keras.layers.Dense(units=num_hidden_units))\n", " model.add(tf.keras.layers.LeakyReLU())\n", " model.add(tf.keras.layers.Dropout(rate=0.5))\n", "\n", " model.add(\n", " tf.keras.layers.Dense(\n", " units=num_output_units,\n", " activation=None)\n", " )\n", " return model" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 247 }, "id": "9Z4JQb9MxD5p", "outputId": "958cf770-b114-4cbb-ac25-c23e3a5d6457" }, "outputs": [ { "data": { "text/html": [ "
Model: \"sequential\"\n",
       "
\n" ], "text/plain": [ "\u001b[1mModel: \"sequential\"\u001b[0m\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (Dense)                   │ (None, 100)            │         2,000 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ leaky_re_lu (LeakyReLU)         │ (None, 100)            │             0 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (Dense)                 │ (None, 784)            │        79,184 │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "
\n" ], "text/plain": [ "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n", "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n", "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m) │ \u001b[38;5;34m2,000\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ leaky_re_lu (\u001b[38;5;33mLeakyReLU\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m784\u001b[0m) │ \u001b[38;5;34m79,184\u001b[0m │\n", "└─────────────────────────────────┴────────────────────────┴───────────────┘\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Total params: 81,184 (317.12 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m81,184\u001b[0m (317.12 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Trainable params: 81,184 (317.12 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m81,184\u001b[0m (317.12 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Non-trainable params: 0 (0.00 B)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "image_size = (28, 28)\n", "z_size = 20\n", "mode_z = 'uniform' # 'uniform' vs. 'normal'\n", "gen_hidden_layers = 1\n", "gen_hidden_size = 100\n", "disc_hidden_layers = 1\n", "disc_hidden_size = 100\n", "\n", "tf.random.set_seed(1)\n", "\n", "gen_model = make_generator_network(\n", " num_hidden_layers=gen_hidden_layers,\n", " num_hidden_units=gen_hidden_size,\n", " num_output_units=np.prod(image_size))\n", "\n", "gen_model.build(input_shape=(None, z_size))\n", "gen_model.summary()" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 283 }, "id": "QAhbOUZHxN1b", "outputId": "184860d5-1e1e-4905-dfdc-331cd0b37d6c" }, "outputs": [ { "data": { "text/html": [ "
Model: \"sequential_1\"\n",
       "
\n" ], "text/plain": [ "\u001b[1mModel: \"sequential_1\"\u001b[0m\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense_2 (Dense)                 │ (None, 100)            │        78,500 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ leaky_re_lu_1 (LeakyReLU)       │ (None, 100)            │             0 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dropout (Dropout)               │ (None, 100)            │             0 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (Dense)                 │ (None, 1)              │           101 │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "
\n" ], "text/plain": [ "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n", "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n", "│ dense_2 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m) │ \u001b[38;5;34m78,500\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ leaky_re_lu_1 (\u001b[38;5;33mLeakyReLU\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dense_3 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m101\u001b[0m │\n", "└─────────────────────────────────┴────────────────────────┴───────────────┘\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Total params: 78,601 (307.04 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m78,601\u001b[0m (307.04 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Trainable params: 78,601 (307.04 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m78,601\u001b[0m (307.04 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Non-trainable params: 0 (0.00 B)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "disc_model = make_discriminator_network(\n", " num_hidden_layers=disc_hidden_layers,\n", " num_hidden_units=disc_hidden_size)\n", "\n", "disc_model.build(input_shape=(None, np.prod(image_size)))\n", "disc_model.summary()" ] }, { "cell_type": "markdown", "metadata": { "id": "edHUQPQxM-OK" }, "source": [ "## 훈련 데이터셋 정의하기" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 245, "referenced_widgets": [ "258b04528a9344439fa8716fe45398c3", "ededc9a8a16e47d3881a70cce82fec76", "7ec2fbbd626d4f1b969e04873730d938", "a778031498174942b17e22d58d5a3153", "1cd4ad1c1e874bb389d1d6b9272e9ecf", "4220d3639ae944b79602f542a436644f", "6e147821ba664bef95e62f9046928ad0", "6142aa12ebb94d5eb19e3a6d4c5d804a", "e0b2ff5d7d26418ca94fed6c252e993f", "bedba5e5bd9f45e386150206c164c27a", "d791bd82b14c4036a469f646edf7f508", "5d2ac8acbe64424296d8537d77279274", "34409533addc488280f75e72793f6f3a", "d3fa856042974c1f909fae22a4d12606", "54a948e9428a4499adf4312c29615cd0", "5422c4abbb7d434daeb54f3448a844b6", "68f98dcfc935490bbf1def4232201943", "844bfc2ba6274f01b99e22b3a15dd984", "243e7cedfdae4ad4afc8d55cac7e33fe", "05c46a4c597f46d789b7e274e8b95c7c", "8a0f30aefab04a1e909676d4cc725f27", "c2bb517fb7514564ad2b8df3275f1df8", "55d88c0d45c848e0ab1deddf116f8146", "ff40548c99134bdf8bb81e6a6aee30d2", "1eaccc8251db4514968b33422091ef6e", "5fecf797356b4206822e567994c60fb9", "114a23ed4d1c42559486440964554200", "36669cea555640d9bdfed6d72678e60d", "b0914a4e11ed4d039d98cec8fd3a7cfb", "9eb6eac230c94d86a883c375f7f01a94", "446e1a30946f4af19e1f51ffb426e094", "a5014117ae78481a962cfc30fbb45778", "6e729a4a4f8343fd9f4f95dca3b2b2f1", "ae53af4e5ea64266a7d25adc3c045488", "4041b53d132b4353a8b12fd0978a055a", "b4140ac190cc42a3af0069a649263500", "e9b9011282d64b1c8fa650d775bf94f6", "aa65f911f4154d71987f8a7a4bfb56d4", "4ac1a810053147d889f0556bd87ac2d5", "1d8cedcd5dc949c4a86d12f26ea933d3", "ea21b9c88c09429abd4a6184e839edf4", "d1f1ecd67e7e4e219ff6cc4f7c7c5c5e", "1473013ac4ca4b5298e9ac194b004353", "56c7e2b7b8f6424783e352faf2a073ba", "16a1d04f6da349abb322dd0f666e7147", "9a87c4b36f73405585d292b8448596b9", "27767ad4498e4abdb02c486dc52e37f5", "7a23e8ef537848d1ac2ac87c8c7b5e47", "aaf8f188a1df4d2a9efe6da7b5d8dc45", "64769257125c400a801287f1b385ce77", "47018c6d589344c9b1f773e5729cc784", "f436121c06b04c7e8f0c8c504787c992", "3e4760a41c3b4882adb8366e594cc8ff", "a37d3189e2d0459eb9413e68efbef5e5", "e9c49143061c4b679e3601d8f33596c6", "bbafbb74755f418c8ceea184a399cf25", "fd22c08926dd481bae79b3c151ebaea8", "696c64625dd54f9ebab4e09ddb666407", "08a08f92fcab4b408a9c033079fbe0e2", "0b1ac652fae34d6ca2c1940c859eb612", "1e6e4e051f7348309bac710172ebc7ec", "6db8bf181d874feab58beff45f2e013a", "1a835923db694f338fa3dcbb6e7a09e4", "88a6aeba7b7e47a6957994d511440ad5", "ba643377aa644041b567d10737e62f12", "3f9b85c75b894260ada8411fdc9e815d", "10a87c36a92c40af8f508d51ba730caa", "a1616cd394f0439093557a967f9d8a2f", "478119408f044f1abb4cc58177af9866", "182b9c80f0664c53988321369342f487", "c355a8c1cca74899898d4034a745228e", "e4f9db5b9bc34c18bc647df672098f1c", "df70910cfbe64e8fb06432444d7d9850", "599d3a081b3642999807d70d0cd3a604", "0b3f2d8862f445bf8a1c9689b7659090", "60f34062f6f942e59d1d49b78c393390", "76847dc0cdfa4f4faf554c48ed20876a", "012572c3c8674d2ca85d8bffdd0255d7", "22d644287f8a46079b59408e7e22af63", "86c80ccb11d34dfcb4867ca055089494", "49d3b66488ee40eca51105dd63db78ae", "36d31e4ab57d480a976f4c1b0c250ad4", "7f7de8104d4148fda0e0a86e8c454eda", "c529c92c4dca4c2b8810f34ee8a7e354", "a294a4add58b407fb5ddebf88f93b1fd", "63e3373fca444d4fad05081ae17e6212", "70b3bae84f154552bd2a66c05ac26daa", "ad7846f4b80a477abbc0940a2d92ea53" ] }, "id": "ApQ1ICJixf2w", "outputId": "c886ed02-6bd6-4f51-91fb-4ebf077d9df4" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Downloading and preparing dataset Unknown size (download: Unknown size, generated: Unknown size, total: Unknown size) to /root/tensorflow_datasets/mnist/3.0.1...\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "258b04528a9344439fa8716fe45398c3", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Dl Completed...: 0 url [00:00, ? url/s]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "5d2ac8acbe64424296d8537d77279274", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Dl Size...: 0 MiB [00:00, ? MiB/s]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "55d88c0d45c848e0ab1deddf116f8146", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Extraction completed...: 0 file [00:00, ? file/s]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "ae53af4e5ea64266a7d25adc3c045488", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Generating splits...: 0%| | 0/2 [00:00 최소: 0 최대: 255\n", "전처리 후: \n", "dtype: 최소: -0.8737728595733643 최대: 0.9460210800170898\n" ] } ], "source": [ "mnist_bldr = tfds.builder('mnist')\n", "mnist_bldr.download_and_prepare()\n", "mnist = mnist_bldr.as_dataset(shuffle_files=False)\n", "\n", "def preprocess(ex, mode='uniform'):\n", " image = ex['image']\n", " image = tf.image.convert_image_dtype(image, tf.float32)\n", " image = tf.reshape(image, [-1])\n", " image = image*2 - 1.0\n", " if mode == 'uniform':\n", " input_z = tf.random.uniform(\n", " shape=(z_size,), minval=-1.0, maxval=1.0)\n", " elif mode == 'normal':\n", " input_z = tf.random.normal(shape=(z_size,))\n", " return input_z, image\n", "\n", "\n", "\n", "mnist_trainset = mnist['train']\n", "\n", "print('전처리 전: ')\n", "example = next(iter(mnist_trainset))['image']\n", "print('dtype: ', example.dtype, ' 최소: {} 최대: {}'.format(np.min(example), np.max(example)))\n", "\n", "mnist_trainset = mnist_trainset.map(preprocess)\n", "\n", "print('전처리 후: ')\n", "example = next(iter(mnist_trainset))[0]\n", "print('dtype: ', example.dtype, ' 최소: {} 최대: {}'.format(np.min(example), np.max(example)))" ] }, { "cell_type": "markdown", "metadata": { "id": "HT57CGAz0RDr" }, "source": [ " * **데이터 흐름을 단계별로 밟아보기**" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "kdAXXUtryFGs", "outputId": "effe6b6c-2543-4375-d67a-5d2286005a37" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "input-z -- 크기: (32, 20)\n", "input-real -- 크기: (32, 784)\n", "생성자 출력 -- 크기: (32, 784)\n", "판별자 (진짜) -- 크기: (32, 1)\n", "판별자 (가짜) -- 크기: (32, 1)\n" ] } ], "source": [ "mnist_trainset = mnist_trainset.batch(32, drop_remainder=True)\n", "input_z, input_real = next(iter(mnist_trainset))\n", "print('input-z -- 크기:', input_z.shape)\n", "print('input-real -- 크기:', input_real.shape)\n", "\n", "g_output = gen_model(input_z)\n", "print('생성자 출력 -- 크기:', g_output.shape)\n", "\n", "d_logits_real = disc_model(input_real)\n", "d_logits_fake = disc_model(g_output)\n", "print('판별자 (진짜) -- 크기:', d_logits_real.shape)\n", "print('판별자 (가짜) -- 크기:', d_logits_fake.shape)" ] }, { "cell_type": "markdown", "metadata": { "id": "Nb2H-Ac3M-OK" }, "source": [ "## GAN 모델 훈련하기" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "-c9xVvjJySZk", "outputId": "64edb9ef-4417-4d34-8d74-dcd11a0a6e28" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "생성자 손실: 0.7089\n", "판별자 손실: 진짜 0.8130 가짜 0.6815\n" ] } ], "source": [ "loss_fn = tf.keras.losses.BinaryCrossentropy(from_logits=True)\n", "\n", "## 생성자 손실\n", "g_labels_real = tf.ones_like(d_logits_fake)\n", "g_loss = loss_fn(y_true=g_labels_real, y_pred=d_logits_fake)\n", "print('생성자 손실: {:.4f}'.format(g_loss))\n", "\n", "## 판별자 손실\n", "d_labels_real = tf.ones_like(d_logits_real)\n", "d_labels_fake = tf.zeros_like(d_logits_fake)\n", "\n", "d_loss_real = loss_fn(y_true=d_labels_real, y_pred=d_logits_real)\n", "d_loss_fake = loss_fn(y_true=d_labels_fake, y_pred=d_logits_fake)\n", "print('판별자 손실: 진짜 {:.4f} 가짜 {:.4f}'\n", " .format(d_loss_real.numpy(), d_loss_fake.numpy()))" ] }, { "cell_type": "markdown", "metadata": { "id": "jtxdTVyF0KCF" }, "source": [ " * **최종 훈련**" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "yeFKLGNfAF5J", "outputId": "94c1435d-5ec0-44d4-a17b-3aa7a3b8956b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "에포크 001 | 시간 1.35 min | 평균 손실 >> 생성자/판별자 3.2017/0.2815 [판별자-진짜: 0.0285 판별자-가짜: 0.2530]\n", "에포크 002 | 시간 2.65 min | 평균 손실 >> 생성자/판별자 5.3511/0.3082 [판별자-진짜: 0.0924 판별자-가짜: 0.2158]\n", "에포크 003 | 시간 3.95 min | 평균 손실 >> 생성자/판별자 3.4550/0.6483 [판별자-진짜: 0.2925 판별자-가짜: 0.3558]\n", "에포크 004 | 시간 5.25 min | 평균 손실 >> 생성자/판별자 2.2002/0.7970 [판별자-진짜: 0.4196 판별자-가짜: 0.3774]\n", "에포크 005 | 시간 6.54 min | 평균 손실 >> 생성자/판별자 2.2485/0.7534 [판별자-진짜: 0.4106 판별자-가짜: 0.3428]\n", "에포크 006 | 시간 7.84 min | 평균 손실 >> 생성자/판별자 2.0143/0.8239 [판별자-진짜: 0.4610 판별자-가짜: 0.3629]\n", "에포크 007 | 시간 9.14 min | 평균 손실 >> 생성자/판별자 1.5902/0.9722 [판별자-진짜: 0.5287 판별자-가짜: 0.4435]\n", "에포크 008 | 시간 10.45 min | 평균 손실 >> 생성자/판별자 1.5996/0.9435 [판별자-진짜: 0.5197 판별자-가짜: 0.4238]\n", "에포크 009 | 시간 11.74 min | 평균 손실 >> 생성자/판별자 1.4642/1.0075 [판별자-진짜: 0.5548 판별자-가짜: 0.4527]\n", "에포크 010 | 시간 13.03 min | 평균 손실 >> 생성자/판별자 1.3118/1.0669 [판별자-진짜: 0.5769 판별자-가짜: 0.4899]\n", "에포크 011 | 시간 14.33 min | 평균 손실 >> 생성자/판별자 1.2982/1.0965 [판별자-진짜: 0.5846 판별자-가짜: 0.5119]\n", "에포크 012 | 시간 15.65 min | 평균 손실 >> 생성자/판별자 1.3332/1.1464 [판별자-진짜: 0.5890 판별자-가짜: 0.5574]\n", "에포크 013 | 시간 16.95 min | 평균 손실 >> 생성자/판별자 1.2801/1.0899 [판별자-진짜: 0.5810 판별자-가짜: 0.5088]\n", "에포크 014 | 시간 18.25 min | 평균 손실 >> 생성자/판별자 1.1983/1.1485 [판별자-진짜: 0.5982 판별자-가짜: 0.5503]\n", "에포크 015 | 시간 19.55 min | 평균 손실 >> 생성자/판별자 1.2948/1.1339 [판별자-진짜: 0.5888 판별자-가짜: 0.5451]\n", "에포크 016 | 시간 20.85 min | 평균 손실 >> 생성자/판별자 1.2189/1.1476 [판별자-진짜: 0.5999 판별자-가짜: 0.5477]\n", "에포크 017 | 시간 22.15 min | 평균 손실 >> 생성자/판별자 1.1447/1.1521 [판별자-진짜: 0.6027 판별자-가짜: 0.5494]\n", "에포크 018 | 시간 23.46 min | 평균 손실 >> 생성자/판별자 1.2051/1.1770 [판별자-진짜: 0.6050 판별자-가짜: 0.5720]\n", "에포크 019 | 시간 24.77 min | 평균 손실 >> 생성자/판별자 1.1601/1.1751 [판별자-진짜: 0.6055 판별자-가짜: 0.5696]\n", "에포크 020 | 시간 26.08 min | 평균 손실 >> 생성자/판별자 1.0895/1.2320 [판별자-진짜: 0.6267 판별자-가짜: 0.6053]\n", "에포크 021 | 시간 27.38 min | 평균 손실 >> 생성자/판별자 1.1267/1.2290 [판별자-진짜: 0.6224 판별자-가짜: 0.6066]\n", "에포크 022 | 시간 28.69 min | 평균 손실 >> 생성자/판별자 1.0776/1.2088 [판별자-진짜: 0.6183 판별자-가짜: 0.5905]\n", "에포크 023 | 시간 30.00 min | 평균 손실 >> 생성자/판별자 1.1012/1.1989 [판별자-진짜: 0.6128 판별자-가짜: 0.5861]\n", "에포크 024 | 시간 31.31 min | 평균 손실 >> 생성자/판별자 1.0922/1.2316 [판별자-진짜: 0.6218 판별자-가짜: 0.6098]\n", "에포크 025 | 시간 32.62 min | 평균 손실 >> 생성자/판별자 1.0793/1.2265 [판별자-진짜: 0.6226 판별자-가짜: 0.6039]\n", "에포크 026 | 시간 33.94 min | 평균 손실 >> 생성자/판별자 1.0536/1.2244 [판별자-진짜: 0.6255 판별자-가짜: 0.5989]\n", "에포크 027 | 시간 35.25 min | 평균 손실 >> 생성자/판별자 1.0676/1.2150 [판별자-진짜: 0.6199 판별자-가짜: 0.5950]\n", "에포크 028 | 시간 36.56 min | 평균 손실 >> 생성자/판별자 1.0542/1.2198 [판별자-진짜: 0.6190 판별자-가짜: 0.6008]\n", "에포크 029 | 시간 37.88 min | 평균 손실 >> 생성자/판별자 1.0627/1.2345 [판별자-진짜: 0.6247 판별자-가짜: 0.6098]\n", "에포크 030 | 시간 39.20 min | 평균 손실 >> 생성자/판별자 1.0646/1.2267 [판별자-진짜: 0.6209 판별자-가짜: 0.6058]\n", "에포크 031 | 시간 40.51 min | 평균 손실 >> 생성자/판별자 1.0067/1.2337 [판별자-진짜: 0.6261 판별자-가짜: 0.6076]\n", "에포크 032 | 시간 41.83 min | 평균 손실 >> 생성자/판별자 1.0119/1.2565 [판별자-진짜: 0.6346 판별자-가짜: 0.6219]\n", "에포크 033 | 시간 43.15 min | 평균 손실 >> 생성자/판별자 0.9597/1.2875 [판별자-진짜: 0.6483 판별자-가짜: 0.6392]\n", "에포크 034 | 시간 44.47 min | 평균 손실 >> 생성자/판별자 1.0011/1.2760 [판별자-진짜: 0.6443 판별자-가짜: 0.6318]\n", "에포크 035 | 시간 45.78 min | 평균 손실 >> 생성자/판별자 1.0123/1.2686 [판별자-진짜: 0.6386 판별자-가짜: 0.6300]\n", "에포크 036 | 시간 47.10 min | 평균 손실 >> 생성자/판별자 0.9549/1.2813 [판별자-진짜: 0.6462 판별자-가짜: 0.6352]\n", "에포크 037 | 시간 48.42 min | 평균 손실 >> 생성자/판별자 0.9541/1.2818 [판별자-진짜: 0.6466 판별자-가짜: 0.6352]\n", "에포크 038 | 시간 49.73 min | 평균 손실 >> 생성자/판별자 1.0119/1.2688 [판별자-진짜: 0.6377 판별자-가짜: 0.6311]\n", "에포크 039 | 시간 51.05 min | 평균 손실 >> 생성자/판별자 1.0722/1.2354 [판별자-진짜: 0.6208 판별자-가짜: 0.6147]\n", "에포크 040 | 시간 52.37 min | 평균 손실 >> 생성자/판별자 0.9743/1.2629 [판별자-진짜: 0.6378 판별자-가짜: 0.6250]\n", "에포크 041 | 시간 53.68 min | 평균 손실 >> 생성자/판별자 0.9415/1.2833 [판별자-진짜: 0.6463 판별자-가짜: 0.6370]\n", "에포크 042 | 시간 55.00 min | 평균 손실 >> 생성자/판별자 1.0145/1.2774 [판별자-진짜: 0.6403 판별자-가짜: 0.6371]\n", "에포크 043 | 시간 56.32 min | 평균 손실 >> 생성자/판별자 1.0383/1.2640 [판별자-진짜: 0.6339 판별자-가짜: 0.6301]\n", "에포크 044 | 시간 57.63 min | 평균 손실 >> 생성자/판별자 0.9655/1.2717 [판별자-진짜: 0.6401 판별자-가짜: 0.6316]\n", "에포크 045 | 시간 58.95 min | 평균 손실 >> 생성자/판별자 0.9397/1.2785 [판별자-진짜: 0.6460 판별자-가짜: 0.6325]\n", "에포크 046 | 시간 60.27 min | 평균 손실 >> 생성자/판별자 0.9714/1.2756 [판별자-진짜: 0.6457 판별자-가짜: 0.6299]\n", "에포크 047 | 시간 61.59 min | 평균 손실 >> 생성자/판별자 0.9898/1.2824 [판별자-진짜: 0.6428 판별자-가짜: 0.6396]\n", "에포크 048 | 시간 62.91 min | 평균 손실 >> 생성자/판별자 0.9542/1.3062 [판별자-진짜: 0.6560 판별자-가짜: 0.6503]\n", "에포크 049 | 시간 64.22 min | 평균 손실 >> 생성자/판별자 0.9112/1.3036 [판별자-진짜: 0.6538 판별자-가짜: 0.6498]\n", "에포크 050 | 시간 65.53 min | 평균 손실 >> 생성자/판별자 0.9977/1.2909 [판별자-진짜: 0.6445 판별자-가짜: 0.6464]\n", "에포크 051 | 시간 66.86 min | 평균 손실 >> 생성자/판별자 0.9861/1.2760 [판별자-진짜: 0.6407 판별자-가짜: 0.6353]\n", "에포크 052 | 시간 68.17 min | 평균 손실 >> 생성자/판별자 0.9238/1.2929 [판별자-진짜: 0.6503 판별자-가짜: 0.6426]\n", "에포크 053 | 시간 69.49 min | 평균 손실 >> 생성자/판별자 0.9838/1.2866 [판별자-진짜: 0.6440 판별자-가짜: 0.6426]\n", "에포크 054 | 시간 70.81 min | 평균 손실 >> 생성자/판별자 0.9648/1.2915 [판별자-진짜: 0.6487 판별자-가짜: 0.6427]\n", "에포크 055 | 시간 72.12 min | 평균 손실 >> 생성자/판별자 0.9433/1.2918 [판별자-진짜: 0.6489 판별자-가짜: 0.6428]\n", "에포크 056 | 시간 73.44 min | 평균 손실 >> 생성자/판별자 0.9101/1.3034 [판별자-진짜: 0.6548 판별자-가짜: 0.6486]\n", "에포크 057 | 시간 74.75 min | 평균 손실 >> 생성자/판별자 0.9387/1.3029 [판별자-진짜: 0.6541 판별자-가짜: 0.6488]\n", "에포크 058 | 시간 76.06 min | 평균 손실 >> 생성자/판별자 0.9915/1.2865 [판별자-진짜: 0.6439 판별자-가짜: 0.6426]\n", "에포크 059 | 시간 77.36 min | 평균 손실 >> 생성자/판별자 0.9222/1.3069 [판별자-진짜: 0.6569 판별자-가짜: 0.6500]\n", "에포크 060 | 시간 78.67 min | 평균 손실 >> 생성자/판별자 0.9321/1.2962 [판별자-진짜: 0.6522 판별자-가짜: 0.6439]\n", "에포크 061 | 시간 79.98 min | 평균 손실 >> 생성자/판별자 0.9269/1.3031 [판별자-진짜: 0.6552 판별자-가짜: 0.6479]\n", "에포크 062 | 시간 81.29 min | 평균 손실 >> 생성자/판별자 0.9431/1.3115 [판별자-진짜: 0.6560 판별자-가짜: 0.6555]\n", "에포크 063 | 시간 82.60 min | 평균 손실 >> 생성자/판별자 0.9665/1.2985 [판별자-진짜: 0.6532 판별자-가짜: 0.6453]\n", "에포크 064 | 시간 83.91 min | 평균 손실 >> 생성자/판별자 0.8634/1.3240 [판별자-진짜: 0.6674 판별자-가짜: 0.6566]\n", "에포크 065 | 시간 85.22 min | 평균 손실 >> 생성자/판별자 1.0017/1.2882 [판별자-진짜: 0.6434 판별자-가짜: 0.6448]\n", "에포크 066 | 시간 86.54 min | 평균 손실 >> 생성자/판별자 0.9851/1.2890 [판별자-진짜: 0.6456 판별자-가짜: 0.6434]\n", "에포크 067 | 시간 87.85 min | 평균 손실 >> 생성자/판별자 0.8728/1.3203 [판별자-진짜: 0.6643 판별자-가짜: 0.6560]\n", "에포크 068 | 시간 89.16 min | 평균 손실 >> 생성자/판별자 0.9104/1.3181 [판별자-진짜: 0.6601 판별자-가짜: 0.6580]\n", "에포크 069 | 시간 90.48 min | 평균 손실 >> 생성자/판별자 0.9489/1.3065 [판별자-진짜: 0.6535 판별자-가짜: 0.6530]\n", "에포크 070 | 시간 91.80 min | 평균 손실 >> 생성자/판별자 0.9014/1.3186 [판별자-진짜: 0.6625 판별자-가짜: 0.6561]\n", "에포크 071 | 시간 93.12 min | 평균 손실 >> 생성자/판별자 0.9126/1.3201 [판별자-진짜: 0.6605 판별자-가짜: 0.6596]\n", "에포크 072 | 시간 94.44 min | 평균 손실 >> 생성자/판별자 0.9422/1.3129 [판별자-진짜: 0.6593 판별자-가짜: 0.6536]\n", "에포크 073 | 시간 95.75 min | 평균 손실 >> 생성자/판별자 0.9471/1.3104 [판별자-진짜: 0.6557 판별자-가짜: 0.6547]\n", "에포크 074 | 시간 97.07 min | 평균 손실 >> 생성자/판별자 0.8853/1.3275 [판별자-진짜: 0.6662 판별자-가짜: 0.6613]\n", "에포크 075 | 시간 98.39 min | 평균 손실 >> 생성자/판별자 0.8909/1.3285 [판별자-진짜: 0.6642 판별자-가짜: 0.6642]\n", "에포크 076 | 시간 99.70 min | 평균 손실 >> 생성자/판별자 0.9238/1.3083 [판별자-진짜: 0.6572 판별자-가짜: 0.6511]\n", "에포크 077 | 시간 101.02 min | 평균 손실 >> 생성자/판별자 0.9273/1.3240 [판별자-진짜: 0.6629 판별자-가짜: 0.6611]\n", "에포크 078 | 시간 102.33 min | 평균 손실 >> 생성자/판별자 0.9155/1.3161 [판별자-진짜: 0.6608 판별자-가짜: 0.6553]\n", "에포크 079 | 시간 103.65 min | 평균 손실 >> 생성자/판별자 0.9030/1.3262 [판별자-진짜: 0.6640 판별자-가짜: 0.6621]\n", "에포크 080 | 시간 104.97 min | 평균 손실 >> 생성자/판별자 0.9200/1.3185 [판별자-진짜: 0.6622 판별자-가짜: 0.6563]\n", "에포크 081 | 시간 106.28 min | 평균 손실 >> 생성자/판별자 0.9214/1.3142 [판별자-진짜: 0.6573 판별자-가짜: 0.6568]\n", "에포크 082 | 시간 107.59 min | 평균 손실 >> 생성자/판별자 0.9043/1.3243 [판별자-진짜: 0.6646 판별자-가짜: 0.6597]\n", "에포크 083 | 시간 108.89 min | 평균 손실 >> 생성자/판별자 0.9324/1.3148 [판별자-진짜: 0.6587 판별자-가짜: 0.6562]\n", "에포크 084 | 시간 110.21 min | 평균 손실 >> 생성자/판별자 0.8986/1.3215 [판별자-진짜: 0.6648 판별자-가짜: 0.6567]\n", "에포크 085 | 시간 111.52 min | 평균 손실 >> 생성자/판별자 0.9191/1.3183 [판별자-진짜: 0.6598 판별자-가짜: 0.6585]\n", "에포크 086 | 시간 112.82 min | 평균 손실 >> 생성자/판별자 0.9112/1.3118 [판별자-진짜: 0.6592 판별자-가짜: 0.6525]\n", "에포크 087 | 시간 114.13 min | 평균 손실 >> 생성자/판별자 0.9056/1.3226 [판별자-진짜: 0.6641 판별자-가짜: 0.6585]\n", "에포크 088 | 시간 115.43 min | 평균 손실 >> 생성자/판별자 0.8990/1.3229 [판별자-진짜: 0.6636 판별자-가짜: 0.6593]\n", "에포크 089 | 시간 116.74 min | 평균 손실 >> 생성자/판별자 0.8867/1.3272 [판별자-진짜: 0.6633 판별자-가짜: 0.6639]\n", "에포크 090 | 시간 118.05 min | 평균 손실 >> 생성자/판별자 0.9438/1.3167 [판별자-진짜: 0.6607 판별자-가짜: 0.6560]\n", "에포크 091 | 시간 119.36 min | 평균 손실 >> 생성자/판별자 0.9290/1.3135 [판별자-진짜: 0.6578 판별자-가짜: 0.6558]\n", "에포크 092 | 시간 120.67 min | 평균 손실 >> 생성자/판별자 0.8893/1.3156 [판별자-진짜: 0.6596 판별자-가짜: 0.6560]\n", "에포크 093 | 시간 121.98 min | 평균 손실 >> 생성자/판별자 0.9055/1.3210 [판별자-진짜: 0.6620 판별자-가짜: 0.6590]\n", "에포크 094 | 시간 123.29 min | 평균 손실 >> 생성자/판별자 0.9115/1.3227 [판별자-진짜: 0.6643 판별자-가짜: 0.6584]\n", "에포크 095 | 시간 124.60 min | 평균 손실 >> 생성자/판별자 0.9483/1.3195 [판별자-진짜: 0.6589 판별자-가짜: 0.6606]\n", "에포크 096 | 시간 125.91 min | 평균 손실 >> 생성자/판별자 0.8652/1.3300 [판별자-진짜: 0.6698 판별자-가짜: 0.6602]\n", "에포크 097 | 시간 127.22 min | 평균 손실 >> 생성자/판별자 0.9562/1.3147 [판별자-진짜: 0.6590 판별자-가짜: 0.6557]\n", "에포크 098 | 시간 128.53 min | 평균 손실 >> 생성자/판별자 0.9206/1.3081 [판별자-진짜: 0.6559 판별자-가짜: 0.6521]\n", "에포크 099 | 시간 129.84 min | 평균 손실 >> 생성자/판별자 0.8965/1.3277 [판별자-진짜: 0.6651 판별자-가짜: 0.6626]\n", "에포크 100 | 시간 131.15 min | 평균 손실 >> 생성자/판별자 0.9189/1.3113 [판별자-진짜: 0.6589 판별자-가짜: 0.6523]\n" ] } ], "source": [ "import time\n", "\n", "\n", "num_epochs = 100\n", "batch_size = 64\n", "image_size = (28, 28)\n", "z_size = 20\n", "mode_z = 'uniform'\n", "gen_hidden_layers = 1\n", "gen_hidden_size = 100\n", "disc_hidden_layers = 1\n", "disc_hidden_size = 100\n", "\n", "tf.random.set_seed(1)\n", "np.random.seed(1)\n", "\n", "\n", "if mode_z == 'uniform':\n", " fixed_z = tf.random.uniform(\n", " shape=(batch_size, z_size),\n", " minval=-1, maxval=1)\n", "elif mode_z == 'normal':\n", " fixed_z = tf.random.normal(\n", " shape=(batch_size, z_size))\n", "\n", "\n", "def create_samples(g_model, input_z):\n", " g_output = g_model(input_z, training=False)\n", " images = tf.reshape(g_output, (batch_size, *image_size))\n", " return (images+1)/2.0\n", "\n", "## 데이터셋 준비\n", "mnist_trainset = mnist['train']\n", "mnist_trainset = mnist_trainset.map(\n", " lambda ex: preprocess(ex, mode=mode_z))\n", "\n", "mnist_trainset = mnist_trainset.shuffle(10000)\n", "mnist_trainset = mnist_trainset.batch(\n", " batch_size, drop_remainder=True)\n", "\n", "## 모델 준비\n", "with tf.device(device_name):\n", " gen_model = make_generator_network(\n", " num_hidden_layers=gen_hidden_layers,\n", " num_hidden_units=gen_hidden_size,\n", " num_output_units=np.prod(image_size))\n", " gen_model.build(input_shape=(None, z_size))\n", "\n", " disc_model = make_discriminator_network(\n", " num_hidden_layers=disc_hidden_layers,\n", " num_hidden_units=disc_hidden_size)\n", " disc_model.build(input_shape=(None, np.prod(image_size)))\n", "\n", "## 손실 함수와 옵티마이저:\n", "loss_fn = tf.keras.losses.BinaryCrossentropy(from_logits=True)\n", "g_optimizer = tf.keras.optimizers.Adam()\n", "d_optimizer = tf.keras.optimizers.Adam()\n", "\n", "all_losses = []\n", "all_d_vals = []\n", "epoch_samples = []\n", "\n", "start_time = time.time()\n", "for epoch in range(1, num_epochs+1):\n", " epoch_losses, epoch_d_vals = [], []\n", " for i,(input_z,input_real) in enumerate(mnist_trainset):\n", "\n", " ## 생성자 손실을 계산합니다\n", " with tf.GradientTape() as g_tape:\n", " g_output = gen_model(input_z)\n", " d_logits_fake = disc_model(g_output, training=True)\n", " labels_real = tf.ones_like(d_logits_fake)\n", " g_loss = loss_fn(y_true=labels_real, y_pred=d_logits_fake)\n", "\n", " # g_loss의 그래디언트를 계산합니다\n", " g_grads = g_tape.gradient(g_loss, gen_model.trainable_variables)\n", "\n", " # 최적화: 그래디언트를 적용합니다\n", " g_optimizer.apply_gradients(\n", " grads_and_vars=zip(g_grads, gen_model.trainable_variables))\n", "\n", " ## 판별자 손실을 계산합니다\n", " with tf.GradientTape() as d_tape:\n", " d_logits_real = disc_model(input_real, training=True)\n", "\n", " d_labels_real = tf.ones_like(d_logits_real)\n", "\n", " d_loss_real = loss_fn(\n", " y_true=d_labels_real, y_pred=d_logits_real)\n", "\n", " d_logits_fake = disc_model(g_output, training=True)\n", " d_labels_fake = tf.zeros_like(d_logits_fake)\n", "\n", " d_loss_fake = loss_fn(\n", " y_true=d_labels_fake, y_pred=d_logits_fake)\n", "\n", " d_loss = d_loss_real + d_loss_fake\n", "\n", " ## d_loss의 그래디언트를 계산합니다\n", " d_grads = d_tape.gradient(d_loss, disc_model.trainable_variables)\n", "\n", " ## 최적화: 그래디언트를 적용합니다\n", " d_optimizer.apply_gradients(\n", " grads_and_vars=zip(d_grads, disc_model.trainable_variables))\n", "\n", " epoch_losses.append(\n", " (g_loss.numpy(), d_loss.numpy(),\n", " d_loss_real.numpy(), d_loss_fake.numpy()))\n", "\n", " d_probs_real = tf.reduce_mean(tf.sigmoid(d_logits_real))\n", " d_probs_fake = tf.reduce_mean(tf.sigmoid(d_logits_fake))\n", " epoch_d_vals.append((d_probs_real.numpy(), d_probs_fake.numpy()))\n", " all_losses.append(epoch_losses)\n", " all_d_vals.append(epoch_d_vals)\n", " print(\n", " '에포크 {:03d} | 시간 {:.2f} min | 평균 손실 >>'\n", " ' 생성자/판별자 {:.4f}/{:.4f} [판별자-진짜: {:.4f} 판별자-가짜: {:.4f}]'\n", " .format(\n", " epoch, (time.time() - start_time)/60,\n", " *list(np.mean(all_losses[-1], axis=0))))\n", " epoch_samples.append(\n", " create_samples(gen_model, fixed_z).numpy())" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 509 }, "id": "TQyQ8deLaHmw", "outputId": "bd356984-7ad8-492e-9c00-1cc5f0138a23" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import itertools\n", "\n", "\n", "fig = plt.figure(figsize=(16, 6))\n", "\n", "## 손실 그래프\n", "ax = fig.add_subplot(1, 2, 1)\n", "g_losses = [item[0] for item in itertools.chain(*all_losses)]\n", "d_losses = [item[1]/2.0 for item in itertools.chain(*all_losses)]\n", "plt.plot(g_losses, label='Generator loss', alpha=0.95)\n", "plt.plot(d_losses, label='Discriminator loss', alpha=0.95)\n", "plt.legend(fontsize=20)\n", "ax.set_xlabel('Iteration', size=15)\n", "ax.set_ylabel('Loss', size=15)\n", "\n", "epochs = np.arange(1, 101)\n", "epoch2iter = lambda e: e*len(all_losses[-1])\n", "epoch_ticks = [1, 20, 40, 60, 80, 100]\n", "newpos = [epoch2iter(e) for e in epoch_ticks]\n", "ax2 = ax.twiny()\n", "ax2.set_xticks(newpos)\n", "ax2.set_xticklabels(epoch_ticks)\n", "ax2.xaxis.set_ticks_position('bottom')\n", "ax2.xaxis.set_label_position('bottom')\n", "ax2.spines['bottom'].set_position(('outward', 60))\n", "ax2.set_xlabel('Epoch', size=15)\n", "ax2.set_xlim(ax.get_xlim())\n", "ax.tick_params(axis='both', which='major', labelsize=15)\n", "ax2.tick_params(axis='both', which='major', labelsize=15)\n", "\n", "## 판별자의 출력\n", "ax = fig.add_subplot(1, 2, 2)\n", "d_vals_real = [item[0] for item in itertools.chain(*all_d_vals)]\n", "d_vals_fake = [item[1] for item in itertools.chain(*all_d_vals)]\n", "plt.plot(d_vals_real, alpha=0.75, label=r'Real: $D(\\mathbf{x})$')\n", "plt.plot(d_vals_fake, alpha=0.75, label=r'Fake: $D(G(\\mathbf{z}))$')\n", "plt.legend(fontsize=20)\n", "ax.set_xlabel('Iteration', size=15)\n", "ax.set_ylabel('Discriminator output', size=15)\n", "\n", "ax2 = ax.twiny()\n", "ax2.set_xticks(newpos)\n", "ax2.set_xticklabels(epoch_ticks)\n", "ax2.xaxis.set_ticks_position('bottom')\n", "ax2.xaxis.set_label_position('bottom')\n", "ax2.spines['bottom'].set_position(('outward', 60))\n", "ax2.set_xlabel('Epoch', size=15)\n", "ax2.set_xlim(ax.get_xlim())\n", "ax.tick_params(axis='both', which='major', labelsize=15)\n", "ax2.tick_params(axis='both', which='major', labelsize=15)\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "iENdPX_gPoJ7", "outputId": "53d00413-df53-4729-8b1b-db57f72f1cac" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "selected_epochs = [1, 2, 4, 10, 50, 100]\n", "fig = plt.figure(figsize=(10, 14))\n", "for i,e in enumerate(selected_epochs):\n", " for j in range(5):\n", " ax = fig.add_subplot(6, 5, i*5+j+1)\n", " ax.set_xticks([])\n", " ax.set_yticks([])\n", " if j == 0:\n", " ax.text(\n", " -0.06, 0.5, 'Epoch {}'.format(e),\n", " rotation=90, size=18, color='red',\n", " horizontalalignment='right',\n", " verticalalignment='center',\n", " transform=ax.transAxes)\n", "\n", " image = epoch_samples[e-1][j]\n", " ax.imshow(image, cmap='gray_r')\n", "\n", "plt.show()" ] } ], "metadata": { "accelerator": "GPU", "colab": { "gpuType": "A100", "machine_shape": "hm", "name": "ch17-basic-GAN.ipynb", "provenance": [] }, "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.5" }, "widgets": { "application/vnd.jupyter.widget-state+json": { "state": {}, "version_major": 2, "version_minor": 0 } } }, "nbformat": 4, "nbformat_minor": 4 }