{
"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": "markdown",
"source": [
"## 程式特色\n",
"\n",
"\n",
"*下載google訓練集\n",
"\n",
"*建立模型於colab\n"
],
"metadata": {
"id": "OOguCPk2nrUR"
}
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 1：準備工具箱 (匯入必要套件)\n",
"\n",
"【本段目的】：\n",
"在開始寫 AI 程式之前，我們必須先從 Python 的資源庫中拿出需要的「工具」。這包含了\n",
"\n",
"1.import os:作業系統操作。用來與電腦作業系統互動。\n",
"\n",
"2.import urllib.request:網址請求與下載。讓 Python 透過網際網路抓取資料\n",
"\n",
"3.import zipfile\n",
"\n",
"4.抓取網路檔案的套件、負責矩陣運算的 numpy\n",
"\n",
"5.核心機器學習框架 tensorflow\n",
"\n",
"6.畫圖的 matplotlib\n",
"\n",
"7.Sequential:「序列式」模型。Keras 中最直覺的模型架構。就像疊漢堡一樣，讓你把神經網路層「一層一層按順序」疊加上去。\n",
"\n",
"8.Adam:優化器（Optimizer）。模型更新學習方向的演算法。\n"
],
"metadata": {
"id": "g0_4_EnYesJB"
}
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "zZnAzvmidRm-",
"outputId": "c6f69594-56b5-42de-a5dc-4d06b962d7db"
},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"✅ 所有必備套件已成功匯入！\n"
]
}
],
"source": [
"import os\n",
"import urllib.request\n",
"import zipfile\n",
"import numpy as np\n",
"import tensorflow as tf\n",
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout, Rescaling\n",
"from tensorflow.keras.optimizers import Adam\n",
"import matplotlib.pyplot as plt\n",
"from google.colab import files # 引入 Colab 上傳檔案的套件\n",
"\n",
"print(\"✅ 所有必備套件已成功匯入！\")"
]
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 2：下載與解壓縮貓狗資料集\n",
"\n",
"【本段目的】：AI 學習需要課本（數據）。這段程式碼會從 Google 的雲端伺服器下載一份已經整理好的貓狗圖片壓縮檔，並將它解壓縮到 Colab 的暫存資料夾 (/tmp) 中，準備供模型讀取。"
],
"metadata": {
"id": "Nh7rNx67e80Z"
}
},
{
"cell_type": "code",
"source": [
"print(\"正在下載貓狗資料集...\")\n",
"url = 'https://storage.googleapis.com/tensorflow-1-public/course2/cats_and_dogs_filtered.zip'\n",
"zip_path = '/tmp/cats_and_dogs_filtered.zip'\n",
"\n",
"# 執行下載\n",
"urllib.request.urlretrieve(url, zip_path)\n",
"\n",
"print(\"正在解壓縮檔案...\")\n",
"with zipfile.ZipFile(zip_path, 'r') as zip_ref:\n",
" zip_ref.extractall('/tmp')\n",
"\n",
"# 設定解壓後的資料夾路徑\n",
"PATH = '/tmp/cats_and_dogs_filtered'\n",
"train_dir = os.path.join(PATH, 'train')\n",
"validation_dir = os.path.join(PATH, 'validation')\n",
"\n",
"print(\"✅ 資料集下載與解壓縮完成！\")\n",
"print(f\"訓練資料路徑: {train_dir}\")\n",
"print(f\"驗證資料路徑: {validation_dir}\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "WSRYtMiZdagA",
"outputId": "ea4613c3-201a-4bee-abbd-7ef95517d968"
},
"execution_count": 2,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"正在下載貓狗資料集...\n",
"正在解壓縮檔案...\n",
"✅ 資料集下載與解壓縮完成！\n",
"訓練資料路徑: /tmp/cats_and_dogs_filtered/train\n",
"驗證資料路徑: /tmp/cats_and_dogs_filtered/validation\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 3：載入影像並轉換為數字矩陣\n",
"\n",
"【本段目的】：電腦看不懂「圖片」，只看得懂「數字」。\n",
"\n",
"1.這個步驟會將資料夾裡大小不一的照片，統一縮放成 150x150 像素，並把它們打包成每次 32 張（Batch Size）的數字矩陣。\n",
"\n",
"2.Batch Size:模型一次要看多少張圖片，才進行一次自我修正（更新權重）」。\n",
"\n",
"3.同時開啟 cache 和 prefetch 功能，可以大幅加速資料讀取效率。"
],
"metadata": {
"id": "jGXVDLa-f-R1"
}
},
{
"cell_type": "code",
"source": [
"BATCH_SIZE = 32\n",
"IMG_SIZE = (150, 150)\n",
"\n",
"print(\"載入訓練資料（供模型學習用）：\")\n",
"train_dataset = tf.keras.utils.image_dataset_from_directory(\n",
" train_dir, shuffle=True, batch_size=BATCH_SIZE, image_size=IMG_SIZE)\n",
"\n",
"print(\"\\n載入驗證資料（供模型考試評估用）：\")\n",
"validation_dataset = tf.keras.utils.image_dataset_from_directory(\n",
" validation_dir, shuffle=True, batch_size=BATCH_SIZE, image_size=IMG_SIZE)\n",
"\n",
"# 效能優化：讓資料讀取更順暢，避免 GPU 等待資料\n",
"AUTOTUNE = tf.data.AUTOTUNE\n",
"train_dataset = train_dataset.cache().prefetch(buffer_size=AUTOTUNE)\n",
"validation_dataset = validation_dataset.cache().prefetch(buffer_size=AUTOTUNE)\n",
"\n",
"print(\"\\n✅ 影像資料已成功轉換為矩陣通道！\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "-OITXkcYdsNo",
"outputId": "72f1120a-77e4-469a-c710-5a1cde18141b"
},
"execution_count": 3,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"載入訓練資料（供模型學習用）：\n",
"Found 2000 files belonging to 2 classes.\n",
"\n",
"載入驗證資料（供模型考試評估用）：\n",
"Found 1000 files belonging to 2 classes.\n",
"\n",
"✅ 影像資料已成功轉換為矩陣通道！\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [],
"metadata": {
"id": "wRCactBggVwJ"
}
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 4：打造 AI 的大腦 (建立 CNN 模型)\n",
"\n",
"【本段目的】：這是影像辨識的核心！我們使用「卷積神經網路 (CNN)」。\n",
"\n",
"Rescaling（數值縮放層）:數值除以255\n",
"\n",
"1.卷積:透過 Conv2D 像放大鏡一樣掃描圖片特徵（如貓耳、狗鼻）\n",
"\n",
"2.池化:用 MaxPooling2D 濃縮特徵\n",
"\n",
"3.最後透過 Dense 層進行思考與分類。\n",
"\n",
"4.這裡還加入了一層 Dropout(0.5)，這能隨機關閉一半的神經元，防止模型「死背」答案（過擬合）。"
],
"metadata": {
"id": "vw9hzWcEgIwR"
}
},
{
"cell_type": "markdown",
"source": [
""
],
"metadata": {
"id": "jzTiw2Mx8mwL"
}
},
{
"cell_type": "markdown",
"source": [
""
],
"metadata": {
"id": "tmqAJGz98xgy"
}
},
{
"cell_type": "markdown",
"source": [
""
],
"metadata": {
"id": "i3QAI95P8BIa"
}
},
{
"cell_type": "code",
"source": [
"model = Sequential([\n",
" # 預處理：將像素值 0~255 縮放到 0~1 之間\n",
" Rescaling(1./255, input_shape=(150, 150, 3)),\n",
"\n",
" # 卷積層與池化層 (提取圖片特徵)\n",
" Conv2D(32, (3, 3), activation='relu'),\n",
" MaxPooling2D(2, 2),\n",
" Conv2D(64, (3, 3), activation='relu'),\n",
" MaxPooling2D(2, 2),\n",
" Conv2D(128, (3, 3), activation='relu'),\n",
" MaxPooling2D(2, 2),\n",
"\n",
" # 攤平層 (將 2D 特徵轉為 1D 陣列)\n",
" Flatten(),\n",
"\n",
" # 全連接層 (進行最終決策)\n",
" Dense(512, activation='relu'),\n",
" Dropout(0.5), # 防止過擬合的神兵利器\n",
"\n",
" # 輸出層 (1個神經元，輸出 0~1 的機率值：0偏貓，1偏狗)\n",
" Dense(1, activation='sigmoid')\n",
"])\n",
"\n",
"print(\"\\n================ 模型架構與層數資訊 ================\")\n",
"model.summary()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 570
},
"id": "FZfDboVpdzb4",
"outputId": "a94e0e66-9ba3-48c3-e22f-620f63aa4781"
},
"execution_count": 4,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"/usr/local/lib/python3.12/dist-packages/keras/src/layers/preprocessing/data_layer.py:95: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
" super().__init__(**kwargs)\n"
]
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"\n",
"================ 模型架構與層數資訊 ================\n"
]
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1mModel: \"sequential\"\u001b[0m\n"
],
"text/html": [
"Model: \"sequential\"\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"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",
"│ rescaling (\u001b[38;5;33mRescaling\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m150\u001b[0m, \u001b[38;5;34m150\u001b[0m, \u001b[38;5;34m3\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m148\u001b[0m, \u001b[38;5;34m148\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m896\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m74\u001b[0m, \u001b[38;5;34m74\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m72\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m18,496\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m36\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m73,856\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m17\u001b[0m, \u001b[38;5;34m17\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m36992\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m18,940,416\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\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;34m1\u001b[0m) │ \u001b[38;5;34m513\u001b[0m │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
],
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
"│ rescaling (Rescaling) │ (None, 150, 150, 3) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d (Conv2D) │ (None, 148, 148, 32) │ 896 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d (MaxPooling2D) │ (None, 74, 74, 32) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_1 (Conv2D) │ (None, 72, 72, 64) │ 18,496 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_1 (MaxPooling2D) │ (None, 36, 36, 64) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ conv2d_2 (Conv2D) │ (None, 34, 34, 128) │ 73,856 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ max_pooling2d_2 (MaxPooling2D) │ (None, 17, 17, 128) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ flatten (Flatten) │ (None, 36992) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense (Dense) │ (None, 512) │ 18,940,416 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dropout (Dropout) │ (None, 512) │ 0 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense_1 (Dense) │ (None, 1) │ 513 │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m19,034,177\u001b[0m (72.61 MB)\n"
],
"text/html": [
" Total params: 19,034,177 (72.61 MB)\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m19,034,177\u001b[0m (72.61 MB)\n"
],
"text/html": [
" Trainable params: 19,034,177 (72.61 MB)\n",
"\n"
]
},
"metadata": {}
},
{
"output_type": "display_data",
"data": {
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
],
"text/html": [
" Non-trainable params: 0 (0.00 B)\n",
"\n"
]
},
"metadata": {}
}
]
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 5：設定學習規則 (編譯模型)\n",
"\n",
"【本段目的】：大腦建好後，要告訴它「怎麼學習」。\n",
"\n",
"1.我們設定學習率（步伐大小）\n",
"\n",
"2.優化器（Adam，負責指引學習方向）\n",
"\n",
"3.以及損失函數（二元交叉熵，用來計算模型猜錯的程度）\n",
"\n",
"4.並要求模型在訓練時隨時回報「正確率 (Accuracy)」。"
],
"metadata": {
"id": "M61jxnyVgb_Z"
}
},
{
"cell_type": "code",
"source": [
"LEARNING_RATE = 0.001\n",
"\n",
"model.compile(\n",
" optimizer=Adam(learning_rate=LEARNING_RATE),\n",
" loss='binary_crossentropy',\n",
" metrics=['accuracy']\n",
")\n",
"\n",
"print(\"✅ 模型編譯完成，已準備好開始學習！\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "nC3uomk0d6dg",
"outputId": "6a7e3fe8-d23c-45d1-b326-c8012321e309"
},
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"✅ 模型編譯完成，已準備好開始學習！\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"# 儲存格 6：正式開始訓練模型\n",
"\n",
"【本段目的】：讓模型實際看圖片學習\n",
"\n",
"1.EPOCHS = 10 代表我們會讓模型把所有的訓練圖片反覆看過 10 遍。\n",
"\n",
"2.訓練過程中的每一次成績，都會被存放在 history 變數中，供我們待會畫圖分析。\n",
"新增區段\n",
"3.(註：執行此段前建議學生開啟 Colab 的 T4 GPU 加速)"
],
"metadata": {
"id": "I1R9C4PcglDp"
}
},
{
"cell_type": "code",
"source": [
"EPOCHS = 10\n",
"\n",
"print(f\"開始訓練模型，共進行 {EPOCHS} 輪 (Epochs)...\")\n",
"history = model.fit(\n",
" train_dataset,\n",
" validation_data=validation_dataset,\n",
" epochs=EPOCHS\n",
")\n",
"\n",
"print(\"\\n✅ 模型訓練完畢！\")"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "FYmSSsSYd_0w",
"outputId": "c89ce513-77d3-4f4c-f2a3-dd67867dbb40"
},
"execution_count": 6,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"開始訓練模型，共進行 10 輪 (Epochs)...\n",
"Epoch 1/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 129ms/step - accuracy: 0.5290 - loss: 0.7695 - val_accuracy: 0.5260 - val_loss: 0.6896\n",
"Epoch 2/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.5970 - loss: 0.6666 - val_accuracy: 0.6390 - val_loss: 0.6605\n",
"Epoch 3/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.6585 - loss: 0.6171 - val_accuracy: 0.6410 - val_loss: 0.6257\n",
"Epoch 4/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.7185 - loss: 0.5581 - val_accuracy: 0.6830 - val_loss: 0.6255\n",
"Epoch 5/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 28ms/step - accuracy: 0.7530 - loss: 0.4904 - val_accuracy: 0.6820 - val_loss: 0.6428\n",
"Epoch 6/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 29ms/step - accuracy: 0.8185 - loss: 0.4007 - val_accuracy: 0.6880 - val_loss: 0.6986\n",
"Epoch 7/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.8575 - loss: 0.3128 - val_accuracy: 0.6910 - val_loss: 0.8876\n",
"Epoch 8/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.8860 - loss: 0.2661 - val_accuracy: 0.6500 - val_loss: 0.9499\n",
"Epoch 9/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.9075 - loss: 0.2229 - val_accuracy: 0.6740 - val_loss: 1.1456\n",
"Epoch 10/10\n",
"\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 27ms/step - accuracy: 0.9230 - loss: 0.1905 - val_accuracy: 0.6730 - val_loss: 1.1696\n",
"\n",
"✅ 模型訓練完畢！\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"儲存格 7：繪製學習成效圖表\n",
"\n",
"【本段目的】：將剛剛訓練過程中的冷冰冰數字，轉換為直觀的折線圖。\n",
"\n",
"1.藍線是訓練正確率（考考古題）\n",
"\n",
"2.橘線是驗證正確率（考新題目）。\n",
"\n",
"3.理想狀態下，兩條線應該要一起平穩上升，這代表模型真的變聰明了。"
],
"metadata": {
"id": "GT3IDnBug25B"
}
},
{
"cell_type": "code",
"source": [
"acc = history.history['accuracy']\n",
"val_acc = history.history['val_accuracy']\n",
"loss = history.history['loss']\n",
"val_loss = history.history['val_loss']\n",
"epochs_range = range(EPOCHS)\n",
"\n",
"plt.figure(figsize=(14, 5))\n",
"\n",
"# 繪製正確率圖表\n",
"plt.subplot(1, 2, 1)\n",
"plt.plot(epochs_range, acc, label='Training Accuracy', marker='o')\n",
"plt.plot(epochs_range, val_acc, label='Validation Accuracy', marker='o')\n",
"plt.legend(loc='lower right')\n",
"plt.title('Training and Validation Accuracy')\n",
"\n",
"# 繪製錯誤率/損失圖表\n",
"plt.subplot(1, 2, 2)\n",
"plt.plot(epochs_range, loss, label='Training Loss', marker='o')\n",
"plt.plot(epochs_range, val_loss, label='Validation Loss', marker='o')\n",
"plt.legend(loc='upper right')\n",
"plt.title('Training and Validation Loss')\n",
"\n",
"plt.show()"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 468
},
"id": "TXG_Tc09eHxC",
"outputId": "63a0f940-eca0-4e13-f3ad-021a4715a0c7"
},
"execution_count": 7,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"image/png": "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\n"
},
"metadata": {}
}
]
},
{
"cell_type": "markdown",
"source": [
"儲存格8:安裝gradio"
],
"metadata": {
"id": "0__Ic9SKba9u"
}
},
{
"cell_type": "code",
"source": [
"!pip install gradio"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "YA1RQhSwbJBG",
"outputId": "fd439390-6c7f-482f-e071-f47de0034384"
},
"execution_count": 8,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Requirement already satisfied: gradio in /usr/local/lib/python3.12/dist-packages (5.50.0)\n",
"Requirement already satisfied: aiofiles<25.0,>=22.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (24.1.0)\n",
"Requirement already satisfied: anyio<5.0,>=3.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (4.13.0)\n",
"Requirement already satisfied: brotli>=1.1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.2.0)\n",
"Requirement already satisfied: fastapi<1.0,>=0.115.2 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.136.1)\n",
"Requirement already satisfied: ffmpy in /usr/local/lib/python3.12/dist-packages (from gradio) (1.0.0)\n",
"Requirement already satisfied: gradio-client==1.14.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.14.0)\n",
"Requirement already satisfied: groovy~=0.1 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.1.2)\n",
"Requirement already satisfied: httpx<1.0,>=0.24.1 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.28.1)\n",
"Requirement already satisfied: huggingface-hub<2.0,>=0.33.5 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.16.1)\n",
"Requirement already satisfied: jinja2<4.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.1.6)\n",
"Requirement already satisfied: markupsafe<4.0,>=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.0.3)\n",
"Requirement already satisfied: numpy<3.0,>=1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.0.2)\n",
"Requirement already satisfied: orjson~=3.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.11.9)\n",
"Requirement already satisfied: packaging in /usr/local/lib/python3.12/dist-packages (from gradio) (26.2)\n",
"Requirement already satisfied: pandas<3.0,>=1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.2.2)\n",
"Requirement already satisfied: pillow<12.0,>=8.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (11.3.0)\n",
"Requirement already satisfied: pydantic<=2.12.3,>=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.12.3)\n",
"Requirement already satisfied: pydub in /usr/local/lib/python3.12/dist-packages (from gradio) (0.25.1)\n",
"Requirement already satisfied: python-multipart>=0.0.18 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.0.29)\n",
"Requirement already satisfied: pyyaml<7.0,>=5.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (6.0.3)\n",
"Requirement already satisfied: ruff>=0.9.3 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.15.14)\n",
"Requirement already satisfied: safehttpx<0.2.0,>=0.1.6 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.1.7)\n",
"Requirement already satisfied: semantic-version~=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.10.0)\n",
"Requirement already satisfied: starlette<1.0,>=0.40.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.52.1)\n",
"Requirement already satisfied: tomlkit<0.14.0,>=0.12.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.13.3)\n",
"Requirement already satisfied: typer<1.0,>=0.12 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.25.1)\n",
"Requirement already satisfied: typing-extensions~=4.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (4.15.0)\n",
"Requirement already satisfied: uvicorn>=0.14.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.47.0)\n",
"Requirement already satisfied: fsspec in /usr/local/lib/python3.12/dist-packages (from gradio-client==1.14.0->gradio) (2025.3.0)\n",
"Requirement already satisfied: websockets<16.0,>=13.0 in /usr/local/lib/python3.12/dist-packages (from gradio-client==1.14.0->gradio) (15.0.1)\n",
"Requirement already satisfied: idna>=2.8 in /usr/local/lib/python3.12/dist-packages (from anyio<5.0,>=3.0->gradio) (3.15)\n",
"Requirement already satisfied: typing-inspection>=0.4.2 in /usr/local/lib/python3.12/dist-packages (from fastapi<1.0,>=0.115.2->gradio) (0.4.2)\n",
"Requirement already satisfied: annotated-doc>=0.0.2 in /usr/local/lib/python3.12/dist-packages (from fastapi<1.0,>=0.115.2->gradio) (0.0.4)\n",
"Requirement already satisfied: certifi in /usr/local/lib/python3.12/dist-packages (from httpx<1.0,>=0.24.1->gradio) (2026.5.20)\n",
"Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.12/dist-packages (from httpx<1.0,>=0.24.1->gradio) (1.0.9)\n",
"Requirement already satisfied: h11>=0.16 in /usr/local/lib/python3.12/dist-packages (from httpcore==1.*->httpx<1.0,>=0.24.1->gradio) (0.16.0)\n",
"Requirement already satisfied: filelock>=3.10.0 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (3.29.0)\n",
"Requirement already satisfied: hf-xet<2.0.0,>=1.4.3 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (1.5.0)\n",
"Requirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (4.67.3)\n",
"Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2.9.0.post0)\n",
"Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2025.2)\n",
"Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2026.2)\n",
"Requirement already satisfied: annotated-types>=0.6.0 in /usr/local/lib/python3.12/dist-packages (from pydantic<=2.12.3,>=2.0->gradio) (0.7.0)\n",
"Requirement already satisfied: pydantic-core==2.41.4 in /usr/local/lib/python3.12/dist-packages (from pydantic<=2.12.3,>=2.0->gradio) (2.41.4)\n",
"Requirement already satisfied: click>=8.2.1 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (8.4.0)\n",
"Requirement already satisfied: shellingham>=1.3.0 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (1.5.4)\n",
"Requirement already satisfied: rich>=13.8.0 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (13.9.4)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.12/dist-packages (from python-dateutil>=2.8.2->pandas<3.0,>=1.0->gradio) (1.17.0)\n",
"Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.12/dist-packages (from rich>=13.8.0->typer<1.0,>=0.12->gradio) (4.2.0)\n",
"Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.12/dist-packages (from rich>=13.8.0->typer<1.0,>=0.12->gradio) (2.20.0)\n",
"Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.12/dist-packages (from markdown-it-py>=2.2.0->rich>=13.8.0->typer<1.0,>=0.12->gradio) (0.1.2)\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"儲存格 9：上傳圖片實戰預測\n",
"\n",
"【本段目的】：最終的驗收時刻！呼叫 Colab 的上傳介面，讓學生可以從自己的電腦上傳一張沒見過的貓或狗圖片，並經過同樣的預處理步驟後，交給我們剛才訓練好的 AI 來預測結果。"
],
"metadata": {
"id": "Q0FccMkThDtR"
}
},
{
"cell_type": "code",
"source": [
"import gradio as gr\n",
"import numpy as np\n",
"import tensorflow as tf\n",
"\n",
"print(\"\\n================ 啟動 AI 貓狗辨識網頁介面 ================\")\n",
"\n",
"# 1. 定義預測函數 (Gradio 會將使用者上傳的圖片傳入這裡)\n",
"def predict_cat_dog(img):\n",
" if img is None:\n",
" return None\n",
"\n",
" # 將 Gradio 傳入的影像調整為模型訓練時的大小 150x150\n",
" img_resized = tf.image.resize(img, [150, 150])\n",
"\n",
" # 增加一個維度 (因為模型預設接收整批資料 [batch_size, height, width, channels])\n",
" x = np.expand_dims(img_resized, axis=0)\n",
"\n",
" # 進行預測\n",
" predictions = model.predict(x)\n",
" score = predictions[0][0]\n",
"\n",
" # 回傳字典格式，讓 Gradio 自動產生漂亮的「機率條」\n",
" return {\"狗 (Dog) 🐶\": float(score), \"貓 (Cat) 🐱\": float(1 - score)}\n",
"\n",
"# 2. 建立 Web 介面\n",
"demo = gr.Interface(\n",
" fn=predict_cat_dog, # 綁定剛才寫好的預測函數\n",
" inputs=gr.Image(label=\"請上傳照片\"), # 圖片輸入區塊\n",
" outputs=gr.Label(label=\"AI 預測結果\"), # 結果輸出區塊 (會顯示機率)\n",
" title=\"🐱🐶 智光商工電機電子群 AI 貓狗辨識小幫手\",\n",
" description=\"請上傳或直接拖曳一張貓或狗的照片到下方區域，AI 會立刻為您辨識！\"\n",
")\n",
"\n",
"# 3. 啟動介面\n",
"# share=True 會自動產生一個 72 小時有效的公開網址，可以傳給學員用手機測試！\n",
"demo.launch(share=True, debug=True)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 680
},
"id": "SZXUSfokbVdO",
"outputId": "a591646f-dd3e-41f6-9686-cf8ff6e7df96"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"\n",
"================ 啟動 AI 貓狗辨識網頁介面 ================\n",
"Colab notebook detected. This cell will run indefinitely so that you can see errors and logs. To turn off, set debug=False in launch().\n",
"* Running on public URL: https://55c08d43469409d5f5.gradio.live\n",
"\n",
"This share link expires in 1 week. For free permanent hosting and GPU upgrades, run `gradio deploy` from the terminal in the working directory to deploy to Hugging Face Spaces (https://huggingface.co/spaces)\n"
]
},
{
"output_type": "display_data",
"data": {
"text/plain": [
""
],
"text/html": [
""
]
},
"metadata": {}
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1s/step \n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step\n"
]
}
]
}
]
}