pw04/MLP_raw-300-dropout.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ic-cqm0tbUc3"
   },
   "source": [
    "# Multilayer Perceptron from raw data\n",
    "This notebook will guide you through the use of the `keras` package to train a multilayer perceptron for handwritten digits classification. You are going to use the `mnist` dataset from LeCun et al. 1998"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ND1Cf9lXbUc6"
   },
   "source": [
    "## Loading the packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "Pvq0DhLtbUdE"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-04-30 21:19:37.608263: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-04-30 21:19:38.046583: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from keras.datasets import mnist\n",
    "from keras.layers import Dropout\n",
    "from keras.layers.core import Dense\n",
    "from keras.models import Sequential\n",
    "from keras.optimizers import RMSprop\n",
    "from keras.utils import np_utils, plot_model\n",
    "from matplotlib import pyplot as pl\n",
    "from sklearn import metrics as me\n",
    "\n",
    "from utils import plot_confusion_matrix\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Vy0wRnm2bUde"
   },
   "source": [
    "## Using raw data to train a MLP\n",
    "First load the `mnist` dataset and normalize it to be in the range [0, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "id": "8CKuJwcibUdi",
    "outputId": "62b6c3c9-af3b-4769-c557-cd6c74afcefe"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "60000 train samples\n",
      "10000 test samples\n"
     ]
    }
   ],
   "source": [
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "\n",
    "X_train = X_train.reshape(60000, 784)\n",
    "X_test = X_test.reshape(10000, 784)\n",
    "X_train = X_train.astype('float32')\n",
    "X_test = X_test.astype('float32')\n",
    "X_train /= 255\n",
    "X_test /= 255\n",
    "print(X_train.shape[0], 'train samples')\n",
    "print(X_test.shape[0], 'test samples')\n",
    "\n",
    "n_classes = 10\n",
    "# convert class vectors to binary class matrices\n",
    "Y_train = np_utils.to_categorical(y_train, n_classes)\n",
    "Y_test = np_utils.to_categorical(y_test, n_classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "01XwVeDNbUdv"
   },
   "source": [
    "Create the MLP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "id": "DNgzrBJEbUd0",
    "outputId": "f34f82a1-cd6b-474f-8186-a0c61f81a08c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 300)               235500    \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 300)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 10)                3010      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 238,510\n",
      "Trainable params: 238,510\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-04-30 21:19:38.978354: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:38.988007: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:38.988057: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:38.990310: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:38.990352: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:38.990381: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:39.413996: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:39.414123: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:39.414132: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1722] Could not identify NUMA node of platform GPU id 0, defaulting to 0.  Your kernel may not have been built with NUMA support.\n",
      "2023-04-30 21:19:39.414166: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:982] could not open file to read NUMA node: /sys/bus/pci/devices/0000:0c:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2023-04-30 21:19:39.414187: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1635] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 21568 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3090, pci bus id: 0000:0c:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "model = Sequential([\n",
    "    Dense(300, input_shape=(784,), activation='relu'),\n",
    "    Dropout(0.5),\n",
    "    Dense(n_classes, activation='softmax'),\n",
    "])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(\n",
    "    model,\n",
    "    show_shapes=True,\n",
    "    show_layer_names=True,\n",
    "    show_layer_activations=True,\n",
    "    show_trainable=True,\n",
    "    to_file='mlpr-model.png'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Peiq9GR2bUeN"
   },
   "source": [
    "Define some constants and train the MLP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 374
    },
    "id": "NBt-ReqIbUeR",
    "outputId": "b1e07312-0ddd-4a6c-a52d-c762dda98b1b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      " 30/422 [=>............................] - ETA: 2s - loss: 1.0274 - accuracy: 0.7076"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-04-30 21:19:40.969020: I tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:637] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "422/422 [==============================] - 3s 6ms/step - loss: 0.3813 - accuracy: 0.8876 - val_loss: 0.1510 - val_accuracy: 0.9595\n",
      "Epoch 2/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.1932 - accuracy: 0.9433 - val_loss: 0.1078 - val_accuracy: 0.9700\n",
      "Epoch 3/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.1533 - accuracy: 0.9554 - val_loss: 0.0963 - val_accuracy: 0.9727\n",
      "Epoch 4/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.1291 - accuracy: 0.9612 - val_loss: 0.0843 - val_accuracy: 0.9743\n",
      "Epoch 5/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.1142 - accuracy: 0.9667 - val_loss: 0.0791 - val_accuracy: 0.9767\n",
      "Epoch 6/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.1044 - accuracy: 0.9691 - val_loss: 0.0780 - val_accuracy: 0.9785\n",
      "Epoch 7/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.0959 - accuracy: 0.9715 - val_loss: 0.0748 - val_accuracy: 0.9798\n",
      "Epoch 8/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.0896 - accuracy: 0.9740 - val_loss: 0.0677 - val_accuracy: 0.9822\n",
      "Epoch 9/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.0821 - accuracy: 0.9756 - val_loss: 0.0688 - val_accuracy: 0.9822\n",
      "Epoch 10/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.0783 - accuracy: 0.9771 - val_loss: 0.0703 - val_accuracy: 0.9810\n",
      "Epoch 11/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.0744 - accuracy: 0.9783 - val_loss: 0.0720 - val_accuracy: 0.9812\n",
      "Epoch 12/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.0735 - accuracy: 0.9786 - val_loss: 0.0674 - val_accuracy: 0.9825\n",
      "Epoch 13/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.0686 - accuracy: 0.9794 - val_loss: 0.0627 - val_accuracy: 0.9853\n",
      "Epoch 14/15\n",
      "422/422 [==============================] - 3s 7ms/step - loss: 0.0683 - accuracy: 0.9798 - val_loss: 0.0718 - val_accuracy: 0.9820\n",
      "Epoch 15/15\n",
      "422/422 [==============================] - 3s 6ms/step - loss: 0.0656 - accuracy: 0.9804 - val_loss: 0.0680 - val_accuracy: 0.9830\n"
     ]
    }
   ],
   "source": [
    "batch_size = 128\n",
    "n_epoch = 15\n",
    "\n",
    "model.compile(loss='categorical_crossentropy', optimizer=RMSprop(), metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(X_train, Y_train,\n",
    "                    batch_size=batch_size, epochs=n_epoch,\n",
    "                    verbose=1, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7u_zpHr5bUeb"
   },
   "source": [
    "Show the performance of the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 299
    },
    "id": "PHXi21E1bUef",
    "outputId": "b153b1ab-4795-4704-b1a3-f28d4d636043"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.07160002738237381\n",
      "Test accuracy: 0.9819999933242798\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pl.plot(history.history['loss'], label='Training')\n",
    "pl.plot(history.history['val_loss'], label='Testing')\n",
    "pl.legend()\n",
    "pl.grid()\n",
    "\n",
    "score = model.evaluate(X_test, Y_test, verbose=0)\n",
    "print('Test score:', score[0])\n",
    "print('Test accuracy:', score[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jLuFK6MobUet"
   },
   "source": [
    "Confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "id": "PAjwjjrjbUex",
    "outputId": "c96b6cf1-3bb9-4056-a6ec-0f20921828c3",
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 1ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F-score 0: 0.9892911779704233\n",
      "F-score 1: 0.9911971830985915\n",
      "F-score 2: 0.9816425120772947\n",
      "F-score 3: 0.9812807881773399\n",
      "F-score 4: 0.9832742017232641\n",
      "F-score 5: 0.9787472035794182\n",
      "F-score 6: 0.9827856025039123\n",
      "F-score 7: 0.9764936336924582\n",
      "F-score 8: 0.9767921609076844\n",
      "F-score 9: 0.9770916334661354\n"
     ]
    }
   ],
   "source": [
    "pred = np.argmax(model.predict(X_test), axis=-1)\n",
    "cm = me.confusion_matrix(y_test, pred)\n",
    "\n",
    "plot_confusion_matrix(cm, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9])"
   ]
  }
 ],
 "metadata": {
  "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}