exercise-binary-classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "34f68625",
   "metadata": {
    "papermill": {
     "duration": 0.005298,
     "end_time": "2023-08-08T13:52:34.091532",
     "exception": false,
     "start_time": "2023-08-08T13:52:34.086234",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "**This notebook is an exercise in the [Intro to Deep Learning](https://www.kaggle.com/learn/intro-to-deep-learning) course.  You can reference the tutorial at [this link](https://www.kaggle.com/ryanholbrook/binary-classification).**\n",
    "\n",
    "---\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cda577eb",
   "metadata": {
    "papermill": {
     "duration": 0.004175,
     "end_time": "2023-08-08T13:52:34.100345",
     "exception": false,
     "start_time": "2023-08-08T13:52:34.096170",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Introduction #\n",
    "\n",
    "In this exercise, you'll build a model to predict hotel cancellations with a binary classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6689b3e8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:34.111673Z",
     "iopub.status.busy": "2023-08-08T13:52:34.110866Z",
     "iopub.status.idle": "2023-08-08T13:52:34.152768Z",
     "shell.execute_reply": "2023-08-08T13:52:34.151907Z"
    },
    "lines_to_next_cell": 0,
    "papermill": {
     "duration": 0.050155,
     "end_time": "2023-08-08T13:52:34.155131",
     "exception": false,
     "start_time": "2023-08-08T13:52:34.104976",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_21/3338970720.py:3: MatplotlibDeprecationWarning: The seaborn styles shipped by Matplotlib are deprecated since 3.6, as they no longer correspond to the styles shipped by seaborn. However, they will remain available as 'seaborn-v0_8-<style>'. Alternatively, directly use the seaborn API instead.\n",
      "  plt.style.use('seaborn-whitegrid')\n"
     ]
    }
   ],
   "source": [
    "# Setup plotting\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('seaborn-whitegrid')\n",
    "# Set Matplotlib defaults\n",
    "plt.rc('figure', autolayout=True)\n",
    "plt.rc('axes', labelweight='bold', labelsize='large',\n",
    "       titleweight='bold', titlesize=18, titlepad=10)\n",
    "plt.rc('animation', html='html5')\n",
    "\n",
    "# Setup feedback system\n",
    "from learntools.core import binder\n",
    "binder.bind(globals())\n",
    "from learntools.deep_learning_intro.ex6 import *"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab56b0a2",
   "metadata": {
    "papermill": {
     "duration": 0.004048,
     "end_time": "2023-08-08T13:52:34.163469",
     "exception": false,
     "start_time": "2023-08-08T13:52:34.159421",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "First, load the *Hotel Cancellations* dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "97578dbb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:34.173782Z",
     "iopub.status.busy": "2023-08-08T13:52:34.172975Z",
     "iopub.status.idle": "2023-08-08T13:52:36.795709Z",
     "shell.execute_reply": "2023-08-08T13:52:36.794652Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 2.630455,
     "end_time": "2023-08-08T13:52:36.798189",
     "exception": false,
     "start_time": "2023-08-08T13:52:34.167734",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.compose import make_column_transformer\n",
    "\n",
    "hotel = pd.read_csv('../input/dl-course-data/hotel.csv')\n",
    "\n",
    "X = hotel.copy()\n",
    "y = X.pop('is_canceled')\n",
    "\n",
    "X['arrival_date_month'] = \\\n",
    "    X['arrival_date_month'].map(\n",
    "        {'January':1, 'February': 2, 'March':3,\n",
    "         'April':4, 'May':5, 'June':6, 'July':7,\n",
    "         'August':8, 'September':9, 'October':10,\n",
    "         'November':11, 'December':12}\n",
    "    )\n",
    "\n",
    "features_num = [\n",
    "    \"lead_time\", \"arrival_date_week_number\",\n",
    "    \"arrival_date_day_of_month\", \"stays_in_weekend_nights\",\n",
    "    \"stays_in_week_nights\", \"adults\", \"children\", \"babies\",\n",
    "    \"is_repeated_guest\", \"previous_cancellations\",\n",
    "    \"previous_bookings_not_canceled\", \"required_car_parking_spaces\",\n",
    "    \"total_of_special_requests\", \"adr\",\n",
    "]\n",
    "features_cat = [\n",
    "    \"hotel\", \"arrival_date_month\", \"meal\",\n",
    "    \"market_segment\", \"distribution_channel\",\n",
    "    \"reserved_room_type\", \"deposit_type\", \"customer_type\",\n",
    "]\n",
    "\n",
    "transformer_num = make_pipeline(\n",
    "    SimpleImputer(strategy=\"constant\"), # there are a few missing values\n",
    "    StandardScaler(),\n",
    ")\n",
    "transformer_cat = make_pipeline(\n",
    "    SimpleImputer(strategy=\"constant\", fill_value=\"NA\"),\n",
    "    OneHotEncoder(handle_unknown='ignore'),\n",
    ")\n",
    "\n",
    "preprocessor = make_column_transformer(\n",
    "    (transformer_num, features_num),\n",
    "    (transformer_cat, features_cat),\n",
    ")\n",
    "\n",
    "# stratify - make sure classes are evenlly represented across splits\n",
    "X_train, X_valid, y_train, y_valid = \\\n",
    "    train_test_split(X, y, stratify=y, train_size=0.75)\n",
    "\n",
    "X_train = preprocessor.fit_transform(X_train)\n",
    "X_valid = preprocessor.transform(X_valid)\n",
    "\n",
    "input_shape = [X_train.shape[1]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23caf95b",
   "metadata": {
    "papermill": {
     "duration": 0.004792,
     "end_time": "2023-08-08T13:52:36.807884",
     "exception": false,
     "start_time": "2023-08-08T13:52:36.803092",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 1) Define Model #\n",
    "\n",
    "The model we'll use this time will have both batch normalization and dropout layers. To ease reading we've broken the diagram into blocks, but you can define it layer by layer as usual.\n",
    "\n",
    "Define a model with an architecture given by this diagram:\n",
    "\n",
    "<figure style=\"padding: 1em;\">\n",
    "<img src=\"https://storage.googleapis.com/kaggle-media/learn/images/V04o59Z.png\" width=\"400\" alt=\"Diagram of network architecture: BatchNorm, Dense, BatchNorm, Dropout, Dense, BatchNorm, Dropout, Dense.\">\n",
    "<figcaption style=\"textalign: center; font-style: italic\"><center>Diagram of a binary classifier.</center></figcaption>\n",
    "</figure>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "61694d64",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:36.818974Z",
     "iopub.status.busy": "2023-08-08T13:52:36.818585Z",
     "iopub.status.idle": "2023-08-08T13:52:47.053434Z",
     "shell.execute_reply": "2023-08-08T13:52:47.052532Z"
    },
    "lines_to_next_cell": 2,
    "papermill": {
     "duration": 10.242834,
     "end_time": "2023-08-08T13:52:47.055442",
     "exception": false,
     "start_time": "2023-08-08T13:52:36.812608",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/__init__.py:98: UserWarning: unable to load libtensorflow_io_plugins.so: unable to open file: libtensorflow_io_plugins.so, from paths: ['/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/libtensorflow_io_plugins.so']\n",
      "caused by: ['/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/libtensorflow_io_plugins.so: undefined symbol: _ZN3tsl6StatusC1EN10tensorflow5error4CodeESt17basic_string_viewIcSt11char_traitsIcEENS_14SourceLocationE']\n",
      "  warnings.warn(f\"unable to load libtensorflow_io_plugins.so: {e}\")\n",
      "/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/__init__.py:104: UserWarning: file system plugins are not loaded: unable to open file: libtensorflow_io.so, from paths: ['/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/libtensorflow_io.so']\n",
      "caused by: ['/opt/conda/lib/python3.10/site-packages/tensorflow_io/python/ops/libtensorflow_io.so: undefined symbol: _ZTVN10tensorflow13GcsFileSystemE']\n",
      "  warnings.warn(f\"file system plugins are not loaded: {e}\")\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "parent.postMessage({\"jupyterEvent\": \"custom.exercise_interaction\", \"data\": {\"outcomeType\": 1, \"valueTowardsCompletion\": 0.3333333333333333, \"interactionType\": 1, \"questionType\": 2, \"questionId\": \"1_Q1\", \"learnToolsVersion\": \"0.3.4\", \"failureMessage\": \"\", \"exceptionClass\": \"\", \"trace\": \"\"}}, \"*\")"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style=\"color:#33cc33\">Correct</span>"
      ],
      "text/plain": [
       "Correct"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "# YOUR CODE HERE: define the model given in the diagram\n",
    "model = keras.Sequential([\n",
    "    layers.BatchNormalization(input_shape=input_shape),\n",
    "    layers.Dense(256, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(rate=0.3),\n",
    "    layers.Dense(256, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(rate=0.3),\n",
    "    layers.Dense(1, activation='sigmoid')\n",
    "])\n",
    "\n",
    "# Check your answer\n",
    "q_1.check()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c15252d3",
   "metadata": {
    "papermill": {
     "duration": 0.004589,
     "end_time": "2023-08-08T13:52:47.064886",
     "exception": false,
     "start_time": "2023-08-08T13:52:47.060297",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 2) Add Optimizer, Loss, and Metric #\n",
    "\n",
    "Now compile the model with the Adam optimizer and binary versions of the cross-entropy loss and accuracy metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f8a8560c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:47.076697Z",
     "iopub.status.busy": "2023-08-08T13:52:47.075278Z",
     "iopub.status.idle": "2023-08-08T13:52:47.096408Z",
     "shell.execute_reply": "2023-08-08T13:52:47.095436Z"
    },
    "lines_to_next_cell": 0,
    "papermill": {
     "duration": 0.028866,
     "end_time": "2023-08-08T13:52:47.098401",
     "exception": false,
     "start_time": "2023-08-08T13:52:47.069535",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "parent.postMessage({\"jupyterEvent\": \"custom.exercise_interaction\", \"data\": {\"outcomeType\": 1, \"valueTowardsCompletion\": 0.3333333333333333, \"interactionType\": 1, \"questionType\": 2, \"questionId\": \"2_Q2\", \"learnToolsVersion\": \"0.3.4\", \"failureMessage\": \"\", \"exceptionClass\": \"\", \"trace\": \"\"}}, \"*\")"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style=\"color:#33cc33\">Correct</span>"
      ],
      "text/plain": [
       "Correct"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# YOUR CODE HERE\n",
    "model.compile(\n",
    "    optimizer='adam',\n",
    "    loss='binary_crossentropy',\n",
    "    metrics=['binary_accuracy']\n",
    ")\n",
    "\n",
    "# Check your answer\n",
    "q_2.check()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "99949397",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:47.110445Z",
     "iopub.status.busy": "2023-08-08T13:52:47.109797Z",
     "iopub.status.idle": "2023-08-08T13:52:47.114793Z",
     "shell.execute_reply": "2023-08-08T13:52:47.113957Z"
    },
    "lines_to_next_cell": 0,
    "papermill": {
     "duration": 0.013276,
     "end_time": "2023-08-08T13:52:47.116701",
     "exception": false,
     "start_time": "2023-08-08T13:52:47.103425",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Lines below will give you a hint or solution code\n",
    "#q_2.hint()\n",
    "#q_2.solution()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd1bc4d6",
   "metadata": {
    "papermill": {
     "duration": 0.004804,
     "end_time": "2023-08-08T13:52:47.126411",
     "exception": false,
     "start_time": "2023-08-08T13:52:47.121607",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Finally, run this cell to train the model and view the learning curves. It may run for around 60 to 70 epochs, which could take a minute or two."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c6b15a26",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:52:47.138541Z",
     "iopub.status.busy": "2023-08-08T13:52:47.137159Z",
     "iopub.status.idle": "2023-08-08T13:53:30.678044Z",
     "shell.execute_reply": "2023-08-08T13:53:30.677069Z"
    },
    "papermill": {
     "duration": 43.54876,
     "end_time": "2023-08-08T13:53:30.680089",
     "exception": false,
     "start_time": "2023-08-08T13:52:47.131329",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/200\n",
      "175/175 [==============================] - 7s 9ms/step - loss: 0.4843 - binary_accuracy: 0.7712 - val_loss: 0.4292 - val_binary_accuracy: 0.8006\n",
      "Epoch 2/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.4231 - binary_accuracy: 0.8025 - val_loss: 0.4006 - val_binary_accuracy: 0.8121\n",
      "Epoch 3/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.4094 - binary_accuracy: 0.8081 - val_loss: 0.3923 - val_binary_accuracy: 0.8173\n",
      "Epoch 4/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.4008 - binary_accuracy: 0.8134 - val_loss: 0.3907 - val_binary_accuracy: 0.8192\n",
      "Epoch 5/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3969 - binary_accuracy: 0.8153 - val_loss: 0.3869 - val_binary_accuracy: 0.8207\n",
      "Epoch 6/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3929 - binary_accuracy: 0.8168 - val_loss: 0.3818 - val_binary_accuracy: 0.8230\n",
      "Epoch 7/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3889 - binary_accuracy: 0.8189 - val_loss: 0.3824 - val_binary_accuracy: 0.8223\n",
      "Epoch 8/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3849 - binary_accuracy: 0.8214 - val_loss: 0.3796 - val_binary_accuracy: 0.8247\n",
      "Epoch 9/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3834 - binary_accuracy: 0.8213 - val_loss: 0.3775 - val_binary_accuracy: 0.8248\n",
      "Epoch 10/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3793 - binary_accuracy: 0.8236 - val_loss: 0.3742 - val_binary_accuracy: 0.8260\n",
      "Epoch 11/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3764 - binary_accuracy: 0.8269 - val_loss: 0.3742 - val_binary_accuracy: 0.8268\n",
      "Epoch 12/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3756 - binary_accuracy: 0.8260 - val_loss: 0.3721 - val_binary_accuracy: 0.8290\n",
      "Epoch 13/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3716 - binary_accuracy: 0.8279 - val_loss: 0.3712 - val_binary_accuracy: 0.8283\n",
      "Epoch 14/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3705 - binary_accuracy: 0.8281 - val_loss: 0.3704 - val_binary_accuracy: 0.8292\n",
      "Epoch 15/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3686 - binary_accuracy: 0.8294 - val_loss: 0.3698 - val_binary_accuracy: 0.8296\n",
      "Epoch 16/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3654 - binary_accuracy: 0.8317 - val_loss: 0.3651 - val_binary_accuracy: 0.8329\n",
      "Epoch 17/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3639 - binary_accuracy: 0.8320 - val_loss: 0.3670 - val_binary_accuracy: 0.8301\n",
      "Epoch 18/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3637 - binary_accuracy: 0.8320 - val_loss: 0.3651 - val_binary_accuracy: 0.8312\n",
      "Epoch 19/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3614 - binary_accuracy: 0.8330 - val_loss: 0.3607 - val_binary_accuracy: 0.8339\n",
      "Epoch 20/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3600 - binary_accuracy: 0.8337 - val_loss: 0.3631 - val_binary_accuracy: 0.8342\n",
      "Epoch 21/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3573 - binary_accuracy: 0.8363 - val_loss: 0.3621 - val_binary_accuracy: 0.8347\n",
      "Epoch 22/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3572 - binary_accuracy: 0.8349 - val_loss: 0.3630 - val_binary_accuracy: 0.8333\n",
      "Epoch 23/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3557 - binary_accuracy: 0.8355 - val_loss: 0.3598 - val_binary_accuracy: 0.8358\n",
      "Epoch 24/200\n",
      "175/175 [==============================] - 1s 6ms/step - loss: 0.3542 - binary_accuracy: 0.8372 - val_loss: 0.3604 - val_binary_accuracy: 0.8343\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Accuracy'}>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "early_stopping = keras.callbacks.EarlyStopping(\n",
    "    patience=5,\n",
    "    min_delta=0.001,\n",
    "    restore_best_weights=True,\n",
    ")\n",
    "history = model.fit(\n",
    "    X_train, y_train,\n",
    "    validation_data=(X_valid, y_valid),\n",
    "    batch_size=512,\n",
    "    epochs=200,\n",
    "    callbacks=[early_stopping],\n",
    ")\n",
    "\n",
    "history_df = pd.DataFrame(history.history)\n",
    "history_df.loc[:, ['loss', 'val_loss']].plot(title=\"Cross-entropy\")\n",
    "history_df.loc[:, ['binary_accuracy', 'val_binary_accuracy']].plot(title=\"Accuracy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "222949c5",
   "metadata": {
    "papermill": {
     "duration": 0.04146,
     "end_time": "2023-08-08T13:53:30.764424",
     "exception": false,
     "start_time": "2023-08-08T13:53:30.722964",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# 3) Train and Evaluate #\n",
    "\n",
    "\n",
    "What do you think about the learning curves? Does it look like the model underfit or overfit? Was the cross-entropy loss a good stand-in for accuracy?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1f51f20f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-08-08T13:53:30.850870Z",
     "iopub.status.busy": "2023-08-08T13:53:30.849139Z",
     "iopub.status.idle": "2023-08-08T13:53:30.858184Z",
     "shell.execute_reply": "2023-08-08T13:53:30.857287Z"
    },
    "papermill": {
     "duration": 0.054229,
     "end_time": "2023-08-08T13:53:30.860430",
     "exception": false,
     "start_time": "2023-08-08T13:53:30.806201",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "parent.postMessage({\"jupyterEvent\": \"custom.exercise_interaction\", \"data\": {\"outcomeType\": 1, \"valueTowardsCompletion\": 0.3333333333333333, \"interactionType\": 1, \"questionType\": 4, \"questionId\": \"3_Q3\", \"learnToolsVersion\": \"0.3.4\", \"failureMessage\": \"\", \"exceptionClass\": \"\", \"trace\": \"\"}}, \"*\")"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "<span style=\"color:#33cc33\">Correct:</span> \n",
       "\n",
       "Though we can see the training loss continuing to fall, the early stopping callback prevented any overfitting. Moreover, the accuracy rose at the same rate as the cross-entropy fell, so it appears that minimizing cross-entropy was a good stand-in. All in all, it looks like this training was a success!"
      ],
      "text/plain": [
       "Correct: \n",
       "\n",
       "Though we can see the training loss continuing to fall, the early stopping callback prevented any overfitting. Moreover, the accuracy rose at the same rate as the cross-entropy fell, so it appears that minimizing cross-entropy was a good stand-in. All in all, it looks like this training was a success!"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# View the solution (Run this cell to receive credit!)\n",
    "q_3.check()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ced2dc1e",
   "metadata": {
    "papermill": {
     "duration": 0.042244,
     "end_time": "2023-08-08T13:53:30.944818",
     "exception": false,
     "start_time": "2023-08-08T13:53:30.902574",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Conclusion #\n",
    "\n",
    "Congratulations! You've completed Kaggle's *Introduction to Deep Learning* course!\n",
    "\n",
    "With your new skills you're ready to take on more advanced applications like computer vision and sentiment classification. What would you like to do next?\n",
    "\n",
    "Why not try one of our *Getting Started* competitions?\n",
    "\n",
    "- Classify images with TPUs in [**Petals to the Metal**](https://www.kaggle.com/c/tpu-getting-started)\n",
    "- Create art with GANs in [**I'm Something of a Painter Myself**](https://www.kaggle.com/c/gan-getting-started)\n",
    "- Classify Tweets in [**Real or Not? NLP with Disaster Tweets**](https://www.kaggle.com/c/nlp-getting-started)\n",
    "- Detect contradiction and entailment in [**Contradictory, My Dear Watson**](https://www.kaggle.com/c/contradictory-my-dear-watson)\n",
    "\n",
    "Until next time, Kagglers!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9cc050e",
   "metadata": {
    "papermill": {
     "duration": 0.043467,
     "end_time": "2023-08-08T13:53:31.030603",
     "exception": false,
     "start_time": "2023-08-08T13:53:30.987136",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "---\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "*Have questions or comments? Visit the [course discussion forum](https://www.kaggle.com/learn/intro-to-deep-learning/discussion) to chat with other learners.*"
   ]
  }
 ],
 "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": "3.10.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 69.731673,
   "end_time": "2023-08-08T13:53:33.904329",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2023-08-08T13:52:24.172656",
   "version": "2.4.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}