Day088_HW-撰寫自己的 callbacks 函數.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Work\n",
    "1. 請嘗試寫一個 callback 用來記錄各類別在訓練過程中，對驗證集的 True Positive 與 True Negative"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import keras\n",
    "\n",
    "# Disable GPU\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = keras.datasets.cifar10.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## 資料前處理\n",
    "def preproc_x(x, flatten=True):\n",
    "    x = x / 255.\n",
    "    if flatten:\n",
    "        x = x.reshape((len(x), -1))\n",
    "    return x\n",
    "\n",
    "def preproc_y(y, num_classes=10):\n",
    "    if y.shape[-1] == 1:\n",
    "        y = keras.utils.to_categorical(y, num_classes)\n",
    "    return y    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, y_train = train\n",
    "x_test, y_test = test\n",
    "\n",
    "# Preproc the inputs\n",
    "x_train = preproc_x(x_train)\n",
    "x_test = preproc_x(x_test)\n",
    "\n",
    "# Preprc the outputs\n",
    "y_train = preproc_y(y_train)\n",
    "y_test = preproc_y(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import BatchNormalization\n",
    "def build_mlp(input_shape, output_units=10, num_neurons=[256, 256, 256]):\n",
    "    \"\"\"Code Here\n",
    "    建立你的神經網路\n",
    "    \"\"\"\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## 超參數設定\n",
    "LEARNING_RATE = 1e-3\n",
    "EPOCHS = 25\n",
    "BATCH_SIZE = 1024\n",
    "MOMENTUM = 0.95"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 載入 Callbacks\n",
    "from keras.callbacks import Callback\n",
    "\n",
    "# Record_fp_tp\n",
    "class Record_tp_tn(Callback):\n",
    "    \"\"\"Code Here\n",
    "    撰寫一個紀錄 ture positive 與 true negative 數量的 callbacks    \n",
    "    \"\"\"\n",
    "        \n",
    "rec_tptn = Record_tp_tn()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "model = build_mlp(input_shape=x_train.shape[1:])\n",
    "model.summary()\n",
    "optimizer = keras.optimizers.SGD(lr=LEARNING_RATE, nesterov=True, momentum=MOMENTUM)\n",
    "model.compile(loss=\"categorical_crossentropy\", metrics=[\"accuracy\"], optimizer=optimizer)\n",
    "\n",
    "\"\"\"Code Here\n",
    "將 callbacks 放入模型訓練流程中\n",
    "\"\"\"\n",
    "model.fit()\n",
    "\n",
    "# Collect results\n",
    "train_loss = model.history.history[\"loss\"]\n",
    "valid_loss = model.history.history[\"val_loss\"]\n",
    "train_acc = model.history.history[\"acc\"]\n",
    "valid_acc = model.history.history[\"val_acc\"]\n",
    "\n",
    "\"\"\"Code Here\n",
    "將 tp/tn 從訓練完的模型中取出\n",
    "\"\"\"\n",
    "valid_tp = \n",
    "valid_tn = "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "plt.plot(range(len(train_loss)), train_loss, label=\"train loss\")\n",
    "plt.plot(range(len(valid_loss)), valid_loss, label=\"valid loss\")\n",
    "plt.legend()\n",
    "plt.title(\"Loss\")\n",
    "plt.show()\n",
    "\n",
    "plt.plot(range(len(train_acc)), train_acc, label=\"train accuracy\")\n",
    "plt.plot(range(len(valid_acc)), valid_acc, label=\"valid accuracy\")\n",
    "plt.legend()\n",
    "plt.title(\"Accuracy\")\n",
    "plt.show()\n",
    "\n",
    "plt.plot(range(len(valid_tp)), valid_tp, label=\"valid tp\", color=\"navy\")\n",
    "plt.plot(range(len(valid_tn)), valid_tn, label=\"valid tn\", color=\"red\")\n",
    "plt.legend()\n",
    "plt.title(\"True positives and True Negatives\")\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}