13. CIFAR_10_Image_Classification_using_Deep_Learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "KsPSm-QJ0pDZ",
    "outputId": "256aabab-ba70-4ea9-ee41-d748f6da9b62"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.datasets import cifar10\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "KAxGFpNm0uF7",
    "outputId": "c4c3250a-823c-462e-af12-c92dc4e32a3e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
      "170500096/170498071 [==============================] - 6s 0us/step\n"
     ]
    }
   ],
   "source": [
    "(train_X,train_Y),(test_X,test_Y)=cifar10.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 401
    },
    "colab_type": "code",
    "id": "Fk2lXFgH00Ih",
    "outputId": "c884303c-4ac2-49b8-e773-b45e157db081"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n=6\n",
    "plt.figure(figsize=(20,10))\n",
    "for i in range(n):\n",
    "  plt.subplot(330+1+i)\n",
    "  plt.imshow(train_X[i])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "LoPjH2cM04tA"
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Dropout\n",
    "from keras.layers import Flatten\n",
    "from keras.constraints import maxnorm\n",
    "from keras.optimizers import SGD\n",
    "from keras.layers.convolutional import Conv2D\n",
    "from keras.layers.convolutional import MaxPooling2D\n",
    "from keras.utils import np_utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "etUve_7n1bxI"
   },
   "outputs": [],
   "source": [
    "train_x=train_X.astype('float32')\n",
    "test_X=test_X.astype('float32')\n",
    "\n",
    "train_X=train_X/255.0\n",
    "test_X=test_X/255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BaodgtSj1dJz"
   },
   "outputs": [],
   "source": [
    "train_Y=np_utils.to_categorical(train_Y)\n",
    "test_Y=np_utils.to_categorical(test_Y)\n",
    "\n",
    "num_classes=test_Y.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YIxjNX-e1iDM"
   },
   "outputs": [],
   "source": [
    "model=Sequential()\n",
    "model.add(Conv2D(32,(3,3),input_shape=(32,32,3),\n",
    "padding='same',activation='relu',\n",
    "kernel_constraint=maxnorm(3)))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Conv2D(32,(3,3),activation='relu',padding='same',kernel_constraint=maxnorm(3)))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(512,activation='relu',kernel_constraint=maxnorm(3)))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(num_classes, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Cyq10OQ_1vjr"
   },
   "outputs": [],
   "source": [
    "sgd=SGD(lr=0.01,momentum=0.9,decay=(0.01/25),nesterov=False)\n",
    "\n",
    "model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 433
    },
    "colab_type": "code",
    "id": "_iJJ7Gu119uj",
    "outputId": "679e1e1e-ad1c-4c38-c88b-4479cf7e4060"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 32, 32, 32)        896       \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 32, 32, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 32, 32, 32)        9248      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 16, 16, 32)        0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 8192)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 512)               4194816   \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 4,210,090\n",
      "Trainable params: 4,210,090\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 398
    },
    "colab_type": "code",
    "id": "bpKWazDD2OIZ",
    "outputId": "d54bcd29-f959-40a3-9405-718f48b97d39"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 50000 samples, validate on 10000 samples\n",
      "Epoch 1/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 1.6807 - accuracy: 0.3939 - val_loss: 1.3306 - val_accuracy: 0.5330\n",
      "Epoch 2/10\n",
      "50000/50000 [==============================] - 210s 4ms/step - loss: 1.2949 - accuracy: 0.5356 - val_loss: 1.1631 - val_accuracy: 0.5858\n",
      "Epoch 3/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 1.1200 - accuracy: 0.6017 - val_loss: 1.0515 - val_accuracy: 0.6265\n",
      "Epoch 4/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 0.9864 - accuracy: 0.6482 - val_loss: 1.0066 - val_accuracy: 0.6456\n",
      "Epoch 5/10\n",
      "50000/50000 [==============================] - 212s 4ms/step - loss: 0.8911 - accuracy: 0.6841 - val_loss: 0.9600 - val_accuracy: 0.6585\n",
      "Epoch 6/10\n",
      "50000/50000 [==============================] - 212s 4ms/step - loss: 0.8057 - accuracy: 0.7161 - val_loss: 0.9254 - val_accuracy: 0.6761\n",
      "Epoch 7/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 0.7293 - accuracy: 0.7447 - val_loss: 0.9062 - val_accuracy: 0.6852\n",
      "Epoch 8/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 0.6637 - accuracy: 0.7662 - val_loss: 0.8852 - val_accuracy: 0.6903\n",
      "Epoch 9/10\n",
      "50000/50000 [==============================] - 212s 4ms/step - loss: 0.6073 - accuracy: 0.7869 - val_loss: 0.8901 - val_accuracy: 0.6940\n",
      "Epoch 10/10\n",
      "50000/50000 [==============================] - 211s 4ms/step - loss: 0.5586 - accuracy: 0.8019 - val_loss: 0.8917 - val_accuracy: 0.6987\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x7f609213fc18>"
      ]
     },
     "execution_count": 13,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_X,train_Y,\n",
    "validation_data=(test_X,test_Y),\n",
    "epochs=10,batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "QKTVxoJ22XNn",
    "outputId": "70b38363-ebac-48d8-96ed-dceff189b4d3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 8s 834us/step\n",
      "69.87000107765198\n"
     ]
    }
   ],
   "source": [
    "_,acc=model.evaluate(test_X,test_Y)\n",
    "print(acc*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "h-Saj0-7AE2Q"
   },
   "outputs": [],
   "source": [
    "model.save(\"model1_cifar_10epoch.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-04T06:32:29.729239Z",
     "start_time": "2020-06-04T06:31:52.176510Z"
    },
    "colab": {},
    "colab_type": "code",
    "id": "9TnyDnjJAm-k"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Admin\\Anaconda3\\lib\\site-packages\\tensorflow_core\\python\\ops\\resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n",
      "1 automobile\n"
     ]
    }
   ],
   "source": [
    "results={\n",
    "0:'aeroplane',\n",
    "1:'automobile',\n",
    "2:'bird',\n",
    "3:'cat',\n",
    "4:'deer',\n",
    "5:'dog',\n",
    "6:'frog',\n",
    "7:'horse',\n",
    "8:'ship',\n",
    "9:'truck'\n",
    "}\n",
    "from PIL import Image\n",
    "import numpy as np\n",
    "im=Image.open(\"truck.jpg\")\n",
    "# the input image is required to be in the shape of dataset, i.e (32,32,3)\n",
    "\n",
    "im=im.resize((32,32))\n",
    "im=np.expand_dims(im,axis=0)\n",
    "im=np.array(im)\n",
    "from tensorflow import keras\n",
    "model = keras.models.load_model('model1_cifar_10epoch.h5')\n",
    "pred=model.predict_classes([im])[0]\n",
    "print(pred,results[pred])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-06-04T06:35:47.476448Z",
     "start_time": "2020-06-04T06:35:06.215187Z"
    },
    "colab": {},
    "colab_type": "code",
    "id": "DumHxHxuAYKE"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "automobile\n"
     ]
    }
   ],
   "source": [
    "import tkinter as tk\n",
    "from tkinter import filedialog\n",
    "from tkinter import *\n",
    "from PIL import ImageTk, Image\n",
    "import numpy\n",
    "\n",
    "#load the trained model to classify the images\n",
    "\n",
    "from keras.models import load_model\n",
    "model = load_model('model1_cifar_10epoch.h5')\n",
    "\n",
    "#dictionary to label all the CIFAR-10 dataset classes.\n",
    "\n",
    "classes = { \n",
    "0:'aeroplane',\n",
    "1:'automobile',\n",
    "2:'bird',\n",
    "3:'cat',\n",
    "4:'deer',\n",
    "5:'dog',\n",
    "6:'frog',\n",
    "7:'horse',\n",
    "8:'ship',\n",
    "9:'truck' \n",
    "}\n",
    "\n",
    "#initialise GUI\n",
    "\n",
    "top=tk.Tk()\n",
    "top.geometry('800x600')\n",
    "top.title('Image Classification CIFAR10')\n",
    "top.configure(background='#CDCDCD')\n",
    "label=Label(top,background='#CDCDCD', font=('arial',15,'bold'))\n",
    "sign_image = Label(top)\n",
    "\n",
    "def classify(file_path):\n",
    "    global label_packed\n",
    "    image = Image.open(file_path)\n",
    "    image = image.resize((32,32))\n",
    "    image = numpy.expand_dims(image, axis=0)\n",
    "    image = numpy.array(image)\n",
    "    pred = model.predict_classes([image])[0]\n",
    "    sign = classes[pred]\n",
    "    print(sign)\n",
    "    label.configure(foreground='#011638', text=sign) \n",
    "\n",
    "def show_classify_button(file_path):\n",
    "    classify_b=Button(top,text=\"Classify Image\",\n",
    "    command=lambda: classify(file_path),padx=10,pady=5)\n",
    "    classify_b.configure(background='#364156', foreground='white',\n",
    "    font=('arial',10,'bold'))\n",
    "    classify_b.place(relx=0.79,rely=0.46)\n",
    "\n",
    "def upload_image():\n",
    "    try:\n",
    "        file_path=filedialog.askopenfilename()\n",
    "        uploaded=Image.open(file_path)\n",
    "        uploaded.thumbnail(((top.winfo_width()/2.25),\n",
    "        (top.winfo_height()/2.25)))\n",
    "        im=ImageTk.PhotoImage(uploaded)\n",
    "        sign_image.configure(image=im)\n",
    "        sign_image.image=im\n",
    "        label.configure(text='')\n",
    "        show_classify_button(file_path)\n",
    "    except:\n",
    "        pass\n",
    "\n",
    "upload=Button(top,text=\"Upload an image\",command=upload_image,\n",
    "padx=10,pady=5)\n",
    "\n",
    "upload.configure(background='#364156', foreground='white',\n",
    "font=('arial',10,'bold'))\n",
    "\n",
    "upload.pack(side=BOTTOM,pady=50)\n",
    "sign_image.pack(side=BOTTOM,expand=True)\n",
    "label.pack(side=BOTTOM,expand=True)\n",
    "heading = Label(top, text=\"Image Classification CIFAR10\",pady=20, font=('arial',20,'bold'))\n",
    "\n",
    "heading.configure(background='#CDCDCD',foreground='#364156')\n",
    "heading.pack()\n",
    "top.mainloop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "CIFAR-10 Image Classification using Deep Learning",
   "provenance": []
  },
  "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": "3.7.4"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}