identified PISA_CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "ename": "ImportError",
     "evalue": "cannot import name 'model'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mImportError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-8-cf9e8acd73ab>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      2\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mmatplotlib\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpyplot\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0mplt\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 4\u001b[1;33m \u001b[1;32mfrom\u001b[0m \u001b[0mtensorflow\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mkeras\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mlayers\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodels\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      5\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[0mtensorflow\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mkeras\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0moptimizers\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mAdam\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[0mtensorflow\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mkeras\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdatasets\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mcifar10\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mImportError\u001b[0m: cannot import name 'model'"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from tensorflow.keras import layers, models\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.datasets import cifar10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 加载 CIFAR-10 数据集\n",
    "(X_train, y_train), (X_test, y_test) = cifar10.load_data()\n",
    "\n",
    "# 数据归一化到 [0, 1]\n",
    "X_train, X_test = X_train / 255.0, X_test / 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SEBlock(layers.Layer):\n",
    "    def __init__(self, reduction_ratio=16):\n",
    "        super(SEBlock, self).__init__()\n",
    "        self.reduction_ratio = reduction_ratio\n",
    "    \n",
    "    def build(self, input_shape):\n",
    "        # 输入特征的通道数\n",
    "        channel = input_shape[-1]\n",
    "        \n",
    "        # 全局平均池化\n",
    "        self.global_avg_pool = layers.GlobalAveragePooling2D()\n",
    "        \n",
    "        # 全连接层，用于生成通道权重\n",
    "        self.dense1 = layers.Dense(channel // self.reduction_ratio, activation='relu', use_bias=False)\n",
    "        self.dense2 = layers.Dense(channel, activation='sigmoid', use_bias=False)\n",
    "    def call(self, inputs):\n",
    "        # 计算通道注意力权重\n",
    "        x = self.global_avg_pool(inputs)\n",
    "        x = layers.Reshape((1, 1, -1))(x)  # Reshape为(batch, 1, 1, channels)\n",
    "        \n",
    "        x = self.dense1(x)\n",
    "        x = self.dense2(x)\n",
    "        \n",
    "        # 对原始特征图进行加权\n",
    "        return inputs * x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Space Attention（空间注意力）\n",
    "class SpatialAttention(layers.Layer):\n",
    "    def __init__(self, kernel_size=7):\n",
    "        super(SpatialAttention, self).__init__()\n",
    "        self.kernel_size = kernel_size\n",
    "    \n",
    "    def build(self, input_shape):\n",
    "        # 使用卷积生成空间注意力图\n",
    "        self.conv = layers.Conv2D(1, (self.kernel_size, self.kernel_size), padding='same', activation='sigmoid')\n",
    "    \n",
    "    def call(self, inputs):\n",
    "        # 对特征图进行卷积生成空间注意力图\n",
    "        attention_map = self.conv(inputs)\n",
    "        return inputs * attention_map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 混合注意力模块：结合SE模块和空间注意力模块\n",
    "class HybridAttention(layers.Layer):\n",
    "    def __init__(self, reduction_ratio=16, kernel_size=7):\n",
    "        super(HybridAttention, self).__init__()\n",
    "        self.se_block = SEBlock(reduction_ratio)\n",
    "        self.spatial_attention = SpatialAttention(kernel_size)\n",
    "    \n",
    "    def call(self, inputs):\n",
    "        # 先应用通道注意力\n",
    "        x = self.se_block(inputs)\n",
    "        # 然后应用空间注意力\n",
    "        x = self.spatial_attention(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:AutoGraph could not transform <bound method HybridAttention.call of <__main__.HybridAttention object at 0x000001F38018D5F8>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <bound method HybridAttention.call of <__main__.HybridAttention object at 0x000001F38018D5F8>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING:tensorflow:AutoGraph could not transform <bound method SEBlock.call of <__main__.SEBlock object at 0x000001F38018D6A0>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <bound method SEBlock.call of <__main__.SEBlock object at 0x000001F38018D6A0>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING:tensorflow:AutoGraph could not transform <bound method SpatialAttention.call of <__main__.SpatialAttention object at 0x000001F38018D7B8>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <bound method SpatialAttention.call of <__main__.SpatialAttention object at 0x000001F38018D7B8>> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 30, 30, 32)        896       \n",
      "_________________________________________________________________\n",
      "hybrid_attention (HybridAtte (None, 30, 30, 32)        1697      \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 15, 15, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 13, 13, 64)        18496     \n",
      "_________________________________________________________________\n",
      "hybrid_attention_1 (HybridAt (None, 13, 13, 64)        3649      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 4, 4, 128)         73856     \n",
      "_________________________________________________________________\n",
      "hybrid_attention_2 (HybridAt (None, 4, 4, 128)         8321      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 2, 2, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               65664     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 173,869\n",
      "Trainable params: 173,869\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# 构建模型并加入混合注意力机制\n",
    "def build_cnn_with_hybrid_attention(input_shape=(32, 32, 3)):\n",
    "    model = models.Sequential([\n",
    "        layers.Conv2D(32, (3, 3), activation='relu', input_shape=input_shape),\n",
    "        HybridAttention(),  # 添加混合注意力机制\n",
    "        layers.MaxPooling2D((2, 2)),\n",
    "        \n",
    "        layers.Conv2D(64, (3, 3), activation='relu'),\n",
    "        HybridAttention(),  # 添加混合注意力机制\n",
    "        layers.MaxPooling2D((2, 2)),\n",
    "        \n",
    "        layers.Conv2D(128, (3, 3), activation='relu'),\n",
    "        HybridAttention(),  # 添加混合注意力机制\n",
    "        layers.MaxPooling2D((2, 2)),\n",
    "        \n",
    "        layers.Flatten(),\n",
    "        layers.Dense(128, activation='relu'),\n",
    "        layers.Dense(10, activation='softmax')  # 10分类（CIFAR-10）\n",
    "    ])\n",
    "    return model\n",
    "model = build_cnn_with_hybrid_attention()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 编译模型\n",
    "# 编译模型\n",
    "model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "WARNING:tensorflow:AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x000001F3EE5D91E0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x000001F3EE5D91E0> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "1563/1563 [==============================] - ETA: 0s - loss: 1.6989 - accuracy: 0.3724WARNING:tensorflow:AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x000001F3EF8E8BF8> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x000001F3EF8E8BF8> and will run it as-is.\n",
      "Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.\n",
      "Cause: 'arguments' object has no attribute 'posonlyargs'\n",
      "To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert\n",
      "1563/1563 [==============================] - 32s 18ms/step - loss: 1.6989 - accuracy: 0.3724 - val_loss: 1.4787 - val_accuracy: 0.4698\n",
      "Epoch 2/10\n",
      "1563/1563 [==============================] - 25s 16ms/step - loss: 1.3109 - accuracy: 0.5264 - val_loss: 1.2156 - val_accuracy: 0.5681\n",
      "Epoch 3/10\n",
      "1563/1563 [==============================] - 20s 13ms/step - loss: 1.1379 - accuracy: 0.5943 - val_loss: 1.1554 - val_accuracy: 0.5945\n",
      "Epoch 4/10\n",
      "1563/1563 [==============================] - 21s 13ms/step - loss: 1.0114 - accuracy: 0.6413 - val_loss: 1.0107 - val_accuracy: 0.6466\n",
      "Epoch 5/10\n",
      "1563/1563 [==============================] - 21s 13ms/step - loss: 0.9280 - accuracy: 0.6738 - val_loss: 0.9478 - val_accuracy: 0.6679\n",
      "Epoch 6/10\n",
      "1563/1563 [==============================] - 22s 14ms/step - loss: 0.8543 - accuracy: 0.7005 - val_loss: 0.9280 - val_accuracy: 0.6733\n",
      "Epoch 7/10\n",
      "1563/1563 [==============================] - 22s 14ms/step - loss: 0.7959 - accuracy: 0.7196 - val_loss: 0.9900 - val_accuracy: 0.6646\n",
      "Epoch 8/10\n",
      "1563/1563 [==============================] - 25s 16ms/step - loss: 0.7486 - accuracy: 0.7382 - val_loss: 0.8833 - val_accuracy: 0.6918\n",
      "Epoch 9/10\n",
      "1563/1563 [==============================] - 21s 13ms/step - loss: 0.7080 - accuracy: 0.7514 - val_loss: 0.8907 - val_accuracy: 0.6965\n",
      "Epoch 10/10\n",
      "1563/1563 [==============================] - 26s 17ms/step - loss: 0.6625 - accuracy: 0.7665 - val_loss: 0.8627 - val_accuracy: 0.7071\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 30, 30, 32)        896       \n",
      "_________________________________________________________________\n",
      "hybrid_attention (HybridAtte (None, 30, 30, 32)        1697      \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 15, 15, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 13, 13, 64)        18496     \n",
      "_________________________________________________________________\n",
      "hybrid_attention_1 (HybridAt (None, 13, 13, 64)        3649      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 4, 4, 128)         73856     \n",
      "_________________________________________________________________\n",
      "hybrid_attention_2 (HybridAt (None, 4, 4, 128)         8321      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 2, 2, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               65664     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 173,869\n",
      "Trainable params: 173,869\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 2s 8ms/step - loss: 0.8627 - accuracy: 0.7071\n",
      "Test accuracy: 0.707099974155426\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# 训练模型并保存 history\n",
    "\n",
    "# 训练模型\n",
    "history = model.fit(X_train, y_train, epochs=10, validation_data=(X_test, y_test))\n",
    "model.summary()\n",
    "\n",
    "# 绘制损失函数和准确率曲线\n",
    "plt.figure(figsize=(12, 6))\n",
    "\n",
    "# 损失函数曲线\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(history.history['loss'], label='Training Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.title('Loss Function')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "\n",
    "# 准确率曲线\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "# 评估模型并输出准确率\n",
    "test_loss, test_acc = model.evaluate(X_test, y_test)\n",
    "print(f\"Test accuracy: {test_acc}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "num_images = 5\n",
    "predictions = model.predict(X_test[:num_images])\n",
    "predicted_classes = np.argmax(predictions, axis=1)\n",
    "\n",
    "# CIFAR-10 类别标签\n",
    "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n",
    "\n",
    "# 显示图像与预测结果\n",
    "plt.figure(figsize=(10, 5))\n",
    "for i in range(num_images):\n",
    "    plt.subplot(1, num_images, i + 1)\n",
    "    plt.imshow(X_test[i])\n",
    "    plt.title(f\"Pred: {class_names[predicted_classes[i]]}\")\n",
    "    plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py36_tensorflow",
   "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}