{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":3362,"databundleVersionId":31148,"sourceType":"competition"}],"dockerImageVersionId":30627,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\n# import os\n# for dirname, _, filenames in os.walk('/kaggle/input'):\n#     for filename in filenames:\n#         print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2026-10-03T15:16:28.997818Z","iopub.execute_input":"2026-10-03T15:16:28.998112Z","iopub.status.idle":"2026-10-03T15:16:29.322304Z","shell.execute_reply.started":"2026-10-03T15:16:28.998081Z","shell.execute_reply":"2026-10-03T15:16:29.321235Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!python --version","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:29.324365Z","iopub.execute_input":"2026-10-03T15:16:29.325121Z","iopub.status.idle":"2026-10-03T15:16:30.396626Z","shell.execute_reply.started":"2026-10-03T15:16:29.325086Z","shell.execute_reply":"2026-10-03T15:16:30.395498Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!uname --all","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:30.398357Z","iopub.execute_input":"2026-10-03T15:16:30.398736Z","iopub.status.idle":"2026-10-03T15:16:31.452594Z","shell.execute_reply.started":"2026-10-03T15:16:30.398687Z","shell.execute_reply":"2026-10-03T15:16:31.45145Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nprint(os.listdir(\"../input\"))\n\nimport zipfile\n\nwith zipfile.ZipFile(\"/kaggle/input/dogs-vs-cats/test1.zip\",\"r\") as z:\n    z.extractall(\".\")\n    \nwith zipfile.ZipFile(\"/kaggle/input/dogs-vs-cats/train.zip\",\"r\") as z:\n    z.extractall(\".\")","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:16:31.454317Z","iopub.execute_input":"2026-10-03T15:16:31.454685Z","iopub.status.idle":"2026-10-03T15:16:40.636065Z","shell.execute_reply.started":"2026-10-03T15:16:31.454659Z","shell.execute_reply":"2026-10-03T15:16:40.635308Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:40.637096Z","iopub.execute_input":"2026-10-03T15:16:40.637371Z","iopub.status.idle":"2026-10-03T15:16:41.6294Z","shell.execute_reply.started":"2026-10-03T15:16:40.637343Z","shell.execute_reply":"2026-10-03T15:16:41.628018Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -l","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:41.630982Z","iopub.execute_input":"2026-10-03T15:16:41.631294Z","iopub.status.idle":"2026-10-03T15:16:42.659907Z","shell.execute_reply.started":"2026-10-03T15:16:41.631271Z","shell.execute_reply":"2026-10-03T15:16:42.658885Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls test1/ | wc -l","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:42.662801Z","iopub.execute_input":"2026-10-03T15:16:42.663163Z","iopub.status.idle":"2026-10-03T15:16:43.713879Z","shell.execute_reply.started":"2026-10-03T15:16:42.663138Z","shell.execute_reply":"2026-10-03T15:16:43.712955Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls train/ | wc -l","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:43.715284Z","iopub.execute_input":"2026-10-03T15:16:43.715581Z","iopub.status.idle":"2026-10-03T15:16:44.822427Z","shell.execute_reply.started":"2026-10-03T15:16:43.715558Z","shell.execute_reply":"2026-10-03T15:16:44.82109Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"#ls train/","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:44.824171Z","iopub.execute_input":"2026-10-03T15:16:44.824568Z","iopub.status.idle":"2026-10-03T15:16:44.829114Z","shell.execute_reply.started":"2026-10-03T15:16:44.824531Z","shell.execute_reply":"2026-10-03T15:16:44.82818Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -U train | head -n 10","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:44.830252Z","iopub.execute_input":"2026-10-03T15:16:44.830612Z","iopub.status.idle":"2026-10-03T15:16:45.912246Z","shell.execute_reply.started":"2026-10-03T15:16:44.830589Z","shell.execute_reply":"2026-10-03T15:16:45.911122Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -U test1 | head -n 10","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:45.914067Z","iopub.execute_input":"2026-10-03T15:16:45.914518Z","iopub.status.idle":"2026-10-03T15:16:46.982853Z","shell.execute_reply.started":"2026-10-03T15:16:45.91449Z","shell.execute_reply":"2026-10-03T15:16:46.98162Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# plot dog photos from the dogs vs cats dataset\nfrom matplotlib import pyplot\n# from matplotlib import pyplot as plt\nfrom matplotlib.image import imread\n# define location of dataset\nfolder = 'train/'\n# plot first few images\nfor i in range(9):\n\t# define subplot\n\tpyplot.subplot(330 + 1 + i)\n\t# define filename\n\tfilename = folder + 'dog.' + str(i) + '.jpg'\n\t# load image pixels\n\timage = imread(filename)\n\t# plot raw pixel data\n\tpyplot.imshow(image)\n# show the figure\npyplot.show()","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:16:46.984341Z","iopub.execute_input":"2026-10-03T15:16:46.984761Z","iopub.status.idle":"2026-10-03T15:16:48.858222Z","shell.execute_reply.started":"2026-10-03T15:16:46.984699Z","shell.execute_reply":"2026-10-03T15:16:48.857384Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# plot cat photos from the dogs vs cats dataset\nfrom matplotlib import pyplot\nfrom matplotlib.image import imread\n# define location of dataset\nfolder = 'train/'\n# plot first few images\nfor i in range(9):\n\t# define subplot\n\tpyplot.subplot(330 + 1 + i)\n\t# define filename\n\tfilename = folder + 'cat.' + str(i) + '.jpg'\n\t# load image pixels\n\timage = imread(filename)\n\t# plot raw pixel data\n\tpyplot.imshow(image)\n# show the figure\npyplot.show()","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:16:48.859402Z","iopub.execute_input":"2026-10-03T15:16:48.860106Z","iopub.status.idle":"2026-10-03T15:16:49.964431Z","shell.execute_reply.started":"2026-10-03T15:16:48.860081Z","shell.execute_reply":"2026-10-03T15:16:49.963382Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!python --version","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:49.965888Z","iopub.execute_input":"2026-10-03T15:16:49.966285Z","iopub.status.idle":"2026-10-03T15:16:51.01511Z","shell.execute_reply.started":"2026-10-03T15:16:49.966242Z","shell.execute_reply":"2026-10-03T15:16:51.013944Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import tensorflow\nprint(tensorflow.__version__)\n\nimport keras\nprint(keras.__version__)\n\nimport tensorflow.keras\nprint(tensorflow.__version__)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:16:51.016506Z","iopub.execute_input":"2026-10-03T15:16:51.017094Z","iopub.status.idle":"2026-10-03T15:17:01.681403Z","shell.execute_reply.started":"2026-10-03T15:16:51.017065Z","shell.execute_reply":"2026-10-03T15:17:01.680273Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# load dogs vs cats dataset, reshape and save to a new file\nfrom os import listdir\nfrom numpy import asarray\nfrom numpy import save\nfrom keras.preprocessing.image import load_img\nfrom keras.preprocessing.image import img_to_array","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:01.682626Z","iopub.execute_input":"2026-10-03T15:17:01.683335Z","iopub.status.idle":"2026-10-03T15:17:01.688272Z","shell.execute_reply.started":"2026-10-03T15:17:01.6833Z","shell.execute_reply":"2026-10-03T15:17:01.687289Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# define location of dataset\nfolder = 'train/'\nphotos, labels = list(), list()\n# enumerate files in the directory\nfor file in listdir(folder):\n\t# determine class\n\toutput = 0.0\n\tif file.startswith('dog'):\n\t\toutput = 1.0\n\t# load image\n\tphoto = load_img(folder + file, target_size=(200, 200))\n\t# convert to numpy array\n\tphoto = img_to_array(photo)\n\t# store\n\tphotos.append(photo)\n\tlabels.append(output)\n# convert to a numpy arrays\nphotos = asarray(photos)\nlabels = asarray(labels)\nprint(photos.shape, labels.shape)\n# save the reshaped photos\nsave('dogs_vs_cats_photos.npy', photos)\nsave('dogs_vs_cats_labels.npy', labels)","metadata":{"execution":{"iopub.status.busy":"2024-01-27T03:36:15.416611Z","iopub.execute_input":"2024-01-27T03:36:15.417128Z","iopub.status.idle":"2024-01-27T03:36:38.886396Z","shell.execute_reply.started":"2024-01-27T03:36:15.417101Z","shell.execute_reply":"2024-01-27T03:36:38.883149Z"}}},{"cell_type":"markdown","source":"# load and confirm the shape\nfrom numpy import load\nphotos = load('dogs_vs_cats_photos.npy')\nlabels = load('dogs_vs_cats_labels.npy')","metadata":{"execution":{"iopub.status.busy":"2024-01-27T03:36:38.88711Z","iopub.status.idle":"2024-01-27T03:36:38.887458Z","shell.execute_reply.started":"2024-01-27T03:36:38.887281Z","shell.execute_reply":"2024-01-27T03:36:38.887296Z"}}},{"cell_type":"code","source":"# organize dataset into a useful structure\nfrom os import makedirs\nfrom os import listdir\nfrom shutil import copyfile\nfrom random import seed\nfrom random import random","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:01.689515Z","iopub.execute_input":"2026-10-03T15:17:01.689978Z","iopub.status.idle":"2026-10-03T15:17:01.70625Z","shell.execute_reply.started":"2026-10-03T15:17:01.689944Z","shell.execute_reply":"2026-10-03T15:17:01.705439Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# create directories\ndataset_home = 'dataset_dogs_vs_cats/'\nsubdirs = ['train/', 'test/'] #test==>validation\nfor subdir in subdirs:\n    # create label subdirectories\n    labeldirs = ['dogs/', 'cats/']\n    for labldir in labeldirs:\n        newdir = dataset_home + subdir + labldir       \n        makedirs(newdir, exist_ok=True)\n        print(newdir)","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:01.707511Z","iopub.execute_input":"2026-10-03T15:17:01.707897Z","iopub.status.idle":"2026-10-03T15:17:01.716695Z","shell.execute_reply.started":"2026-10-03T15:17:01.707864Z","shell.execute_reply":"2026-10-03T15:17:01.715608Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -l","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:01.717829Z","iopub.execute_input":"2026-10-03T15:17:01.718046Z","iopub.status.idle":"2026-10-03T15:17:02.746931Z","shell.execute_reply.started":"2026-10-03T15:17:01.718026Z","shell.execute_reply":"2026-10-03T15:17:02.745628Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -l dataset_dogs_vs_cats/train","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:02.748504Z","iopub.execute_input":"2026-10-03T15:17:02.748912Z","iopub.status.idle":"2026-10-03T15:17:03.743095Z","shell.execute_reply.started":"2026-10-03T15:17:02.748873Z","shell.execute_reply":"2026-10-03T15:17:03.741807Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls -l dataset_dogs_vs_cats/test #validation","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:03.745676Z","iopub.execute_input":"2026-10-03T15:17:03.746044Z","iopub.status.idle":"2026-10-03T15:17:04.747298Z","shell.execute_reply.started":"2026-10-03T15:17:03.746014Z","shell.execute_reply":"2026-10-03T15:17:04.746071Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# seed random number generator\nseed(1337)\n# define ratio of pictures to use for validation\nval_ratio = 0.25 #changable\n# copy training dataset images into subdirectories\nsrc_directory = 'train/'\n\nfor file in listdir(src_directory):\n\tsrc = src_directory + '/' + file\n\tdst_dir = 'train/'\n\tif random() < val_ratio:\n\t\tdst_dir = 'test/'\n\tif file.startswith('cat'):\n\t\tdst = dataset_home + dst_dir + 'cats/'  + file\n\t\tcopyfile(src, dst)\n\telif file.startswith('dog'):\n\t\tdst = dataset_home + dst_dir + 'dogs/'  + file\n\t\tcopyfile(src, dst)","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:04.753257Z","iopub.execute_input":"2026-10-03T15:17:04.753523Z","iopub.status.idle":"2026-10-03T15:17:06.628607Z","shell.execute_reply.started":"2026-10-03T15:17:04.753498Z","shell.execute_reply":"2026-10-03T15:17:06.627788Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"print(src)\nprint(dst)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:06.629909Z","iopub.execute_input":"2026-10-03T15:17:06.630205Z","iopub.status.idle":"2026-10-03T15:17:06.636287Z","shell.execute_reply.started":"2026-10-03T15:17:06.630181Z","shell.execute_reply":"2026-10-03T15:17:06.635435Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!ls -l dataset_dogs_vs_cats/train/dogs | wc -l\n!ls -l dataset_dogs_vs_cats/train/cats | wc -l","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:06.637454Z","iopub.execute_input":"2026-10-03T15:17:06.638199Z","iopub.status.idle":"2026-10-03T15:17:08.77241Z","shell.execute_reply.started":"2026-10-03T15:17:06.638167Z","shell.execute_reply":"2026-10-03T15:17:08.771282Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"#validation set\n!ls -l dataset_dogs_vs_cats/test/dogs | wc -l\n!ls -l dataset_dogs_vs_cats/test/cats | wc -l","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:08.773831Z","iopub.execute_input":"2026-10-03T15:17:08.774103Z","iopub.status.idle":"2026-10-03T15:17:10.794632Z","shell.execute_reply.started":"2026-10-03T15:17:08.774076Z","shell.execute_reply":"2026-10-03T15:17:10.793662Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!pip -q install livelossplot==0.5.5","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:10.796086Z","iopub.execute_input":"2026-10-03T15:17:10.796484Z","iopub.status.idle":"2026-10-03T15:17:20.5533Z","shell.execute_reply.started":"2026-10-03T15:17:10.796443Z","shell.execute_reply":"2026-10-03T15:17:20.552212Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from livelossplot import PlotLossesKeras","metadata":{"execution":{"iopub.status.busy":"2026-10-03T15:17:20.554831Z","iopub.execute_input":"2026-10-03T15:17:20.555191Z","iopub.status.idle":"2026-10-03T15:17:20.572674Z","shell.execute_reply.started":"2026-10-03T15:17:20.555157Z","shell.execute_reply":"2026-10-03T15:17:20.571641Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Prepare Final Dataset","metadata":{}},{"cell_type":"code","source":"# https://keras.io/api/applications/","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:20.57399Z","iopub.execute_input":"2026-10-03T15:17:20.574296Z","iopub.status.idle":"2026-10-03T15:17:20.57964Z","shell.execute_reply.started":"2026-10-03T15:17:20.574267Z","shell.execute_reply":"2026-10-03T15:17:20.578894Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# baseline model for the dogs vs cats dataset\nimport sys\nfrom matplotlib import pyplot\nfrom keras.utils import to_categorical\nfrom keras.models import Sequential\nfrom keras.layers import Conv2D\nfrom keras.layers import MaxPooling2D\nfrom keras.layers import Dense\nfrom keras.layers import Flatten\nfrom keras.optimizers import SGD\nfrom keras.preprocessing.image import ImageDataGenerator","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:20.580845Z","iopub.execute_input":"2026-10-03T15:17:20.581587Z","iopub.status.idle":"2026-10-03T15:17:20.591199Z","shell.execute_reply.started":"2026-10-03T15:17:20.581564Z","shell.execute_reply":"2026-10-03T15:17:20.590481Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# define cnn model\ndef define_model_1():\n\tmodel = Sequential()\n\tmodel.add(Conv2D(32, (3, 3), activation='relu',\n                     kernel_initializer='he_uniform',\n                     padding='same',\n                     input_shape=(200, 200, 3))\n             )\n\tmodel.add(MaxPooling2D((2, 2)))\n\tmodel.add(Flatten())\n\tmodel.add(Dense(128, activation='relu', kernel_initializer='he_uniform'))\n\tmodel.add(Dense(1, activation='sigmoid'))\n\t# compile model\n\topt = SGD(lr=0.001, momentum=0.9)\n\tmodel.compile(optimizer=opt, loss='binary_crossentropy', metrics=['accuracy'])\n\treturn model","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:20.592167Z","iopub.execute_input":"2026-10-03T15:17:20.592542Z","iopub.status.idle":"2026-10-03T15:17:20.60039Z","shell.execute_reply.started":"2026-10-03T15:17:20.592521Z","shell.execute_reply":"2026-10-03T15:17:20.599554Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model_1 = define_model_1()\nprint(model_1.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:20.601392Z","iopub.execute_input":"2026-10-03T15:17:20.602284Z","iopub.status.idle":"2026-10-03T15:17:22.606529Z","shell.execute_reply.started":"2026-10-03T15:17:20.602252Z","shell.execute_reply":"2026-10-03T15:17:22.605786Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# plot diagnostic learning curves\ndef summarize_diagnostics(history):\n    # plot loss\n    pyplot.subplot(211)\n    pyplot.title('Cross Entropy Loss')\n    pyplot.plot(history.history['loss'], color='blue', label='train')\n    pyplot.plot(history.history['val_loss'], color='orange', label='test')\n    # plot accuracy\n    pyplot.subplot(212)\n    pyplot.title('Classification Accuracy')\n    pyplot.plot(history.history['accuracy'], color='blue', label='train')\n    pyplot.plot(history.history['val_accuracy'], color='orange', label='test')\n    pyplot.show()\n    \n    # save plot to file\n    filename = sys.argv[0].split('/')[-1]\n    pyplot.savefig(filename + '_plot.png') \n    pyplot.close()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.607978Z","iopub.execute_input":"2026-10-03T15:17:22.608424Z","iopub.status.idle":"2026-10-03T15:17:22.614221Z","shell.execute_reply.started":"2026-10-03T15:17:22.608388Z","shell.execute_reply":"2026-10-03T15:17:22.613354Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model = Sequential()\n\nmodel.add(Conv2D(32,\n                 (3, 3),\n                 activation='relu',\n                 input_shape=(200, 200, 3)\n                )\n         )\n\nmodel.add(MaxPooling2D((2, 2)))\n\nmodel.add(Flatten())\n\nmodel.add(Dense(128, activation='relu'))\n\nmodel.add(Dense(2, activation='softmax'))\n\nprint(model.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.615524Z","iopub.execute_input":"2026-10-03T15:17:22.615874Z","iopub.status.idle":"2026-10-03T15:17:22.69561Z","shell.execute_reply.started":"2026-10-03T15:17:22.615843Z","shell.execute_reply":"2026-10-03T15:17:22.694765Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# The Functional API\nfrom tensorflow.keras.models import Model\nfrom tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense\n\n# explicitly define the input tensor topology\nmy_inputs = Input(shape=(200, 200, 3))\n\n# Construct the forward pass of the computational graph\nx1 = Conv2D(32, (3, 3), activation='relu')(my_inputs)\n\nx2 = MaxPooling2D((2, 2))(x1)\n\nx3 = Flatten()(x2)\n\nx5 = Dense(256, activation='relu')(x3)\n\nx4 = Dense(128, activation='relu')(x5)\n\n# Define the output tensor with the target objective function representation\nmy_outputs = Dense(2, activation='softmax')(x4)\n\n# Instantiate the formal model architecture\nmodel = Model(inputs=my_inputs, outputs=my_outputs)\n\nprint(model.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.697179Z","iopub.execute_input":"2026-10-03T15:17:22.697556Z","iopub.status.idle":"2026-10-03T15:17:22.781312Z","shell.execute_reply.started":"2026-10-03T15:17:22.69752Z","shell.execute_reply":"2026-10-03T15:17:22.780385Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# The Functional API - new cell\nfrom tensorflow.keras.models import Model\nfrom tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense\n\n# explicitly define the input tensor topology\nmy_inputs = Input(shape=(200, 200, 3))\n\n# Construct the forward pass of the computational graph\nx1 = Conv2D(32, (3, 3), activation='relu')(my_inputs)\n\nx1 = Conv2D(64, (3, 3), activation='relu')(x1)\n\nx2 = MaxPooling2D((2, 2))(x1)\n\nx3 = Flatten()(x2)\n\nx5 = Dense(256, activation='relu')(x3)\n\nx4 = Dense(128, activation='relu')(x5)\n\n# Define the output tensor with the target objective function representation\nmy_outputs = Dense(2, activation='softmax')(x4)\n\n# Instantiate the formal model architecture\nmodel = Model(inputs=my_inputs, outputs=my_outputs)\n\nprint(model.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.782481Z","iopub.execute_input":"2026-10-03T15:17:22.782878Z","iopub.status.idle":"2026-10-03T15:17:22.876589Z","shell.execute_reply.started":"2026-10-03T15:17:22.782854Z","shell.execute_reply":"2026-10-03T15:17:22.875455Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"datagen = ImageDataGenerator(rescale=1.0/255.0)\ntype(datagen)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.87791Z","iopub.execute_input":"2026-10-03T15:17:22.878184Z","iopub.status.idle":"2026-10-03T15:17:22.885088Z","shell.execute_reply.started":"2026-10-03T15:17:22.878162Z","shell.execute_reply":"2026-10-03T15:17:22.884371Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"ls dataset_dogs_vs_cats/train/","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:22.886039Z","iopub.execute_input":"2026-10-03T15:17:22.886406Z","iopub.status.idle":"2026-10-03T15:17:23.97348Z","shell.execute_reply.started":"2026-10-03T15:17:22.886368Z","shell.execute_reply":"2026-10-03T15:17:23.972487Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"IMAGE_SIZE = 224 #for vgg16\nBATCH_SIZE = 64\nEPOCHS = 30","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:23.976519Z","iopub.execute_input":"2026-10-03T15:17:23.977214Z","iopub.status.idle":"2026-10-03T15:17:23.981937Z","shell.execute_reply.started":"2026-10-03T15:17:23.977186Z","shell.execute_reply":"2026-10-03T15:17:23.9809Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Use another dataset\n\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\n\n# 1. Instantiate the generator with the validation_split ratio\ndatagen = ImageDataGenerator(\n    rescale=1.0/255.0,\n    horizontal_flip=True,\n    validation_split=0.2  # Defined here at initialization\n)\n\ntrain_path = '/kaggle/input/datasets/kasikrit/308-456-dogs-vs-cats-dataset/dataset_dogs_vs_cats/train'\n\n# 2. Extract the training subset (80% of data)\ntrain_it = datagen.flow_from_directory(\n    train_path,\n    class_mode='binary',\n    batch_size=BATCH_SIZE,\n    target_size=(IMAGE_SIZE, IMAGE_SIZE),\n    subset='training',     # Explicitly request the training partition\n    shuffle=True           # Stochastic sampling is standard for training\n)\n\n# 3. Extract the validation subset (20% of data)\nval_it = datagen.flow_from_directory(\n    train_path,\n    class_mode='binary',\n    batch_size=BATCH_SIZE,\n    target_size=(IMAGE_SIZE, IMAGE_SIZE),\n    subset='validation',   # Explicitly request the validation partition\n    shuffle=False          # Disable shuffle to maintain index alignment for metrics\n)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:17:23.983447Z","iopub.execute_input":"2026-10-03T15:17:23.98368Z","iopub.status.idle":"2026-10-03T15:17:50.496543Z","shell.execute_reply.started":"2026-10-03T15:17:23.983654Z","shell.execute_reply":"2026-10-03T15:17:50.49584Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# =============================================================================\n# Dogs vs Cats — data pipeline with a switchable stratified 10% sample\n#\n# Paste each BLOCK into its own notebook cell, in order.\n# To switch corpora, edit exactly one line: DATA_MODE in BLOCK 0.\n# =============================================================================\n\n\n# =============================================================================\n# BLOCK 0 — CONFIG\n# =============================================================================\nimport os\nimport json\nimport math\nimport random\nimport shutil\nimport time\n\nimport numpy as np\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\n\nDATA_MODE       = 'full'   # 'full' | 'sample'   <- the only line you edit\nSAMPLE_FRACTION = 0.15       # ignored when DATA_MODE == 'full'\nVAL_SPLIT       = 0.2\nSEED            = 42\n\nBATCH_SIZE = 32              # keep your existing values\nIMAGE_SIZE = 224\n\nFULL_TRAIN_PATH = ('/kaggle/input/datasets/kasikrit/308-456-dogs-vs-cats-dataset'\n                   '/dataset_dogs_vs_cats/train')\nSAMPLE_ROOT     = '/kaggle/working/sample_train'\nMANIFEST_NAME   = '.sample_manifest.json'\n\nassert DATA_MODE in ('full', 'sample'), f\"DATA_MODE must be 'full' or 'sample', got {DATA_MODE!r}\"\n\n# sorted() matches the alphabetical ordering flow_from_directory uses internally,\n# so passing this as `classes=` locks the mapping to {'cats': 0, 'dogs': 1}.\nCLASS_NAMES = sorted(d for d in os.listdir(FULL_TRAIN_PATH)\n                     if os.path.isdir(os.path.join(FULL_TRAIN_PATH, d)))\n\nprint(f'DATA_MODE      : {DATA_MODE}')\nprint(f'classes        : {CLASS_NAMES}')\nprint(f'seed           : {SEED}')\n\n\n# =============================================================================\n# BLOCK 1 — BUILD THE STRATIFIED SUBSET TREE   (runs only when DATA_MODE == 'sample')\n#\n# Per-class sampling IS the stratification: taking round(N_c * fraction) files from\n# each class directory preserves the original class proportions by construction.\n# The tree is made of symlinks, so it costs no disk and takes seconds. It must live\n# under /kaggle/working because /kaggle/input is read-only.\n# =============================================================================\nIMG_EXT = ('.jpg', '.jpeg', '.png', '.bmp', '.ppm', '.tif', '.tiff', '.gif')\n\n\ndef list_class_files(root, cls):\n    \"\"\"Sorted image filenames for one class, so the listing order is not FS-dependent.\"\"\"\n    d = os.path.join(root, cls)\n    return sorted(f for f in os.listdir(d) if f.lower().endswith(IMG_EXT))\n\n\ndef build_stratified_sample(src_root, dst_root, classes, fraction, seed):\n    manifest_path = os.path.join(dst_root, MANIFEST_NAME)\n    wanted = {'source': src_root, 'fraction': fraction, 'seed': seed, 'classes': list(classes)}\n\n    if os.path.isdir(dst_root):\n        # Safety guard: only ever delete a tree this function created.\n        if not os.path.exists(manifest_path):\n            raise RuntimeError(\n                f'{dst_root} exists but has no {MANIFEST_NAME}; refusing to delete it. '\n                'Point SAMPLE_ROOT somewhere else or remove that directory yourself.')\n        with open(manifest_path) as fh:\n            old = json.load(fh)\n        if all(old.get(k) == v for k, v in wanted.items()):\n            print(f'reusing existing sample at {dst_root}')\n            for cls, c in old['counts'].items():\n                print(f\"  {cls:<10} {c['sampled']:>6} / {c['available']:>6}\")\n            return dst_root\n        print(f'config changed -> rebuilding {dst_root}')\n        shutil.rmtree(dst_root)\n\n    rng = random.Random(seed)          # one stream, so the whole draw is reproducible\n    counts, symlinked = {}, True\n\n    for cls in classes:\n        files = list_class_files(src_root, cls)\n        n = max(1, round(len(files) * fraction))\n        picked = rng.sample(files, n)\n\n        outdir = os.path.join(dst_root, cls)\n        os.makedirs(outdir, exist_ok=True)\n        for fname in picked:\n            # abspath matters: a relative symlink target is resolved relative to the\n            # LINK's directory, not the cwd, which would silently produce dead links.\n            src = os.path.abspath(os.path.join(src_root, cls, fname))\n            dst = os.path.join(outdir, fname)\n            try:\n                os.symlink(src, dst)\n            except (OSError, NotImplementedError, AttributeError):\n                shutil.copy2(src, dst)  # filesystem without symlink support\n                symlinked = False\n\n        counts[cls] = {'sampled': n, 'available': len(files)}\n        print(f'  {cls:<10} {n:>6} / {len(files):>6}  ({n / len(files):.1%})')\n\n    manifest = dict(wanted,\n                    counts=counts,\n                    symlinked=symlinked,\n                    created=time.strftime('%Y-%m-%d %H:%M:%S'))\n    with open(manifest_path, 'w') as fh:\n        json.dump(manifest, fh, indent=2)\n\n    print(f'built {dst_root}  (symlinked={symlinked})')\n    return dst_root\n\n\nif DATA_MODE == 'full':\n    build_stratified_sample(FULL_TRAIN_PATH, SAMPLE_ROOT,\n                            CLASS_NAMES, SAMPLE_FRACTION, SEED)\nelse:\n    print('DATA_MODE=full -> skipping sample build')\n\n\n# =============================================================================\n# BLOCK 2 — SELECT THE ACTIVE CORPUS\n# Everything downstream reads TRAIN_DIR and is identical between the two modes.\n# =============================================================================\nTRAIN_DIR = SAMPLE_ROOT if DATA_MODE == 'sample' else FULL_TRAIN_PATH\nprint(f'TRAIN_DIR = {TRAIN_DIR}')\n\n\n# =============================================================================\n# BLOCK 3 — TWO GENERATORS\n#\n# Augmentation on train only. The original code shared one generator, so validation\n# images were randomly flipped every epoch: val metrics wobbled for no modelling\n# reason and the confusion matrix was computed on augmented inputs.\n#\n# Both generators MUST carry the same validation_split. flow_from_directory\n# partitions by slicing the sorted filename list *within each class directory*, so\n# the 80/20 split is itself stratified, is deterministic, and is unaffected by\n# `seed` — two separate instances therefore yield identical, non-overlapping\n# partitions. BLOCK 5 asserts this rather than trusting it.\n# =============================================================================\ntrain_datagen = ImageDataGenerator(\n    rescale=1.0 / 255.0,\n    rotation_range=20,        # [เพิ่มเข้ามา] สุ่มหมุนภาพไม่เกิน 20 องศา\n    width_shift_range=0.2,    # [เพิ่มเข้ามา] เลื่อนภาพซ้ายขวา 20%\n    height_shift_range=0.2,   # [เพิ่มเข้ามา] เลื่อนภาพขึ้นลง 20%\n    zoom_range=0.2,           # [เพิ่มเข้ามา] สุ่มซูมภาพ 20%\n    horizontal_flip=True,     # [ของเดิมที่มีอยู่แล้ว]\n    validation_split=VAL_SPLIT, # [ของเดิมที่มีอยู่แล้ว] ห้ามลบนะครับ เพราะมันใช้แบ่งข้อมูล\n)\n\nval_datagen = ImageDataGenerator(\n    rescale=1.0 / 255.0,          # no augmentation (ถูกต้องแล้ว ห้ามใส่ลูกเล่นในข้อมูล Test/Val)\n    validation_split=VAL_SPLIT,\n)\n\n\n# =============================================================================\n# BLOCK 4 — THE TWO ITERATORS\n# =============================================================================\ntrain_it = train_datagen.flow_from_directory(\n    TRAIN_DIR,\n    classes=CLASS_NAMES,          # lock the label mapping explicitly\n    class_mode='binary',\n    batch_size=BATCH_SIZE,\n    target_size=(IMAGE_SIZE, IMAGE_SIZE),\n    subset='training',\n    shuffle=True,\n    seed=SEED,\n)\n\nval_it = val_datagen.flow_from_directory(\n    TRAIN_DIR,\n    classes=CLASS_NAMES,\n    class_mode='binary',\n    batch_size=BATCH_SIZE,\n    target_size=(IMAGE_SIZE, IMAGE_SIZE),\n    subset='validation',\n    shuffle=False,                # keep index alignment for the confusion matrix\n)\n\n\n# =============================================================================\n# BLOCK 5 — VERIFICATION (print + assert)\n# =============================================================================\ndef class_counts(it):\n    n = np.bincount(it.classes, minlength=len(CLASS_NAMES))\n    return {cls: int(c) for cls, c in zip(CLASS_NAMES, n)}\n\n\nfull_total   = sum(len(list_class_files(FULL_TRAIN_PATH, c)) for c in CLASS_NAMES)\nactive_total = sum(len(list_class_files(TRAIN_DIR, c)) for c in CLASS_NAMES)\n\nprint(f'class_indices  : {train_it.class_indices}')\nprint(f'full train set : {full_total}')\nprint(f'active corpus  : {active_total}  ({active_total / full_total:.1%} of full)')\nprint(f'train          : {train_it.samples:>6}  {class_counts(train_it)}')\nprint(f'val            : {val_it.samples:>6}  {class_counts(val_it)}')\n\nassert train_it.class_indices == {c: i for i, c in enumerate(CLASS_NAMES)}\nassert val_it.class_indices == train_it.class_indices\nassert not (set(train_it.filenames) & set(val_it.filenames)), 'train/val leakage'\nassert train_it.samples + val_it.samples == active_total, 'iterators lost files'\n\n# class balance must survive the sampling\nfor cls, c in class_counts(train_it).items():\n    share = c / train_it.samples\n    assert abs(share - 1 / len(CLASS_NAMES)) < 0.05, f'{cls} is imbalanced in train: {share:.1%}'\n\nprint('all checks passed')\n\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T21:11:16.744879Z","iopub.execute_input":"2026-10-03T21:11:16.74547Z","iopub.status.idle":"2026-10-03T21:11:48.150481Z","shell.execute_reply.started":"2026-10-03T21:11:16.745438Z","shell.execute_reply":"2026-10-03T21:11:48.149686Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"len(train_it), len(train_it.classes), train_it.classes[:10]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.385612Z","iopub.execute_input":"2026-10-03T15:21:31.385965Z","iopub.status.idle":"2026-10-03T15:21:31.393503Z","shell.execute_reply.started":"2026-10-03T15:21:31.385943Z","shell.execute_reply":"2026-10-03T15:21:31.392587Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"len(val_it), len(val_it.classes), val_it.classes[-10:]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.394863Z","iopub.execute_input":"2026-10-03T15:21:31.395271Z","iopub.status.idle":"2026-10-03T15:21:31.40283Z","shell.execute_reply.started":"2026-10-03T15:21:31.395235Z","shell.execute_reply":"2026-10-03T15:21:31.40197Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# fit model\n# model_1.reset_states()\n# history_1 = model_1.fit(\n#     train_it,\n#     steps_per_epoch=len(train_it),\n#     validation_data=val_it,\n#     validation_steps=len(val_it),\n#     epochs=EPOCHS,\n#     verbose=2,\n#     callbacks = [PlotLossesKeras()]\n#     )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.404053Z","iopub.execute_input":"2026-10-03T15:21:31.404336Z","iopub.status.idle":"2026-10-03T15:21:31.418113Z","shell.execute_reply.started":"2026-10-03T15:21:31.404314Z","shell.execute_reply":"2026-10-03T15:21:31.417249Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# learning curves\n# summarize_diagnostics(history_1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.420535Z","iopub.execute_input":"2026-10-03T15:21:31.420835Z","iopub.status.idle":"2026-10-03T15:21:31.428442Z","shell.execute_reply.started":"2026-10-03T15:21:31.420814Z","shell.execute_reply":"2026-10-03T15:21:31.427682Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# evaluate model\n# _, acc = model_1.evaluate(val_it, steps=len(val_it), verbose=1)\n# print('> %.3f' % (acc * 100.0))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.429529Z","iopub.execute_input":"2026-10-03T15:21:31.429852Z","iopub.status.idle":"2026-10-03T15:21:31.43577Z","shell.execute_reply.started":"2026-10-03T15:21:31.429824Z","shell.execute_reply":"2026-10-03T15:21:31.434927Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## Create test datagenerator from 308-456 Dogs vs Cats ","metadata":{}},{"cell_type":"code","source":"ls /kaggle/input/datasets/kasikrit/308-456-dogs-vs-cats-dataset/dataset_dogs_vs_cats/test","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:31.436934Z","iopub.execute_input":"2026-10-03T15:21:31.437397Z","iopub.status.idle":"2026-10-03T15:21:32.507645Z","shell.execute_reply.started":"2026-10-03T15:21:31.437371Z","shell.execute_reply":"2026-10-03T15:21:32.506688Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nimport pandas as pd\nfrom glob import glob\n\n# Define the image folder path\ntest_path = '/kaggle/input/datasets/kasikrit/308-456-dogs-vs-cats-dataset/dataset_dogs_vs_cats/test'\n\nprint(test_path)\n# # List all .jpg files in the directory (adjust the pattern if needed)\n# image_files = glob(os.path.join(image_folder, '*.jpg'))\n\n# # Check if images were found\n# if not image_files:\n#     print(\"No images found. Please verify the directory and file extensions.\")\n    \n# # Create a DataFrame with the file paths\n# test_df = pd.DataFrame({'filename': image_files})\n\n# If your use-case doesn't require labels (for example, during testing), you can leave out the label column.\n# Otherwise, if you need a dummy label, you can add one:\n# df['class'] = 'dummy_label'\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.509022Z","iopub.execute_input":"2026-10-03T15:21:32.509682Z","iopub.status.idle":"2026-10-03T15:21:32.515537Z","shell.execute_reply.started":"2026-10-03T15:21:32.509654Z","shell.execute_reply":"2026-10-03T15:21:32.514328Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"\n","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# test_df.columns","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.516821Z","iopub.execute_input":"2026-10-03T15:21:32.51767Z","iopub.status.idle":"2026-10-03T15:21:32.533752Z","shell.execute_reply.started":"2026-10-03T15:21:32.517645Z","shell.execute_reply":"2026-10-03T15:21:32.532779Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# image_files[:5], type(image_files), len(image_files)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.534911Z","iopub.execute_input":"2026-10-03T15:21:32.535211Z","iopub.status.idle":"2026-10-03T15:21:32.544993Z","shell.execute_reply.started":"2026-10-03T15:21:32.535182Z","shell.execute_reply":"2026-10-03T15:21:32.543647Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# test_df.describe()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.54749Z","iopub.execute_input":"2026-10-03T15:21:32.548096Z","iopub.status.idle":"2026-10-03T15:21:32.556193Z","shell.execute_reply.started":"2026-10-03T15:21:32.548074Z","shell.execute_reply":"2026-10-03T15:21:32.555407Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# test_df.loc[1]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.557188Z","iopub.execute_input":"2026-10-03T15:21:32.557857Z","iopub.status.idle":"2026-10-03T15:21:32.568484Z","shell.execute_reply.started":"2026-10-03T15:21:32.557825Z","shell.execute_reply":"2026-10-03T15:21:32.56777Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# test_df.sample(5)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.569612Z","iopub.execute_input":"2026-10-03T15:21:32.569929Z","iopub.status.idle":"2026-10-03T15:21:32.576452Z","shell.execute_reply.started":"2026-10-03T15:21:32.569901Z","shell.execute_reply":"2026-10-03T15:21:32.575774Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# # Create test data generator\n# from tensorflow.keras.preprocessing.image import ImageDataGenerator\n\n# # Initialize your ImageDataGenerator (modify parameters as needed)\n# datagen_test = ImageDataGenerator(rescale=1./255,\n#                                  vertical_flip=True)\n\n# # Create the data generator using the DataFrame\n# test_it = datagen_test.flow_from_dataframe(\n#     dataframe=test_df,\n#     x_col='filename',\n#     y_col=None,            # No labels for testing, so set to None\n#     target_size=IMAGE_SIZE,\n#     batch_size=BATCH_SIZE,\n#     class_mode=None,       # Since there are no labels\n#     # shuffle=True,         # Usually testing data should not be shuffled\n#     shuffle=False,\n# )\n\n# type(test_it)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.581244Z","iopub.execute_input":"2026-10-03T15:21:32.581605Z","iopub.status.idle":"2026-10-03T15:21:32.588444Z","shell.execute_reply.started":"2026-10-03T15:21:32.581573Z","shell.execute_reply":"2026-10-03T15:21:32.587745Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"IMAGE_SIZE, BATCH_SIZE","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.589541Z","iopub.execute_input":"2026-10-03T15:21:32.589869Z","iopub.status.idle":"2026-10-03T15:21:32.602408Z","shell.execute_reply.started":"2026-10-03T15:21:32.589841Z","shell.execute_reply":"2026-10-03T15:21:32.601647Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# 1. Instantiate the generator with the validation_split ratio\ntest_datagen = ImageDataGenerator(\n    rescale=1.0/255.0,\n    horizontal_flip=True\n)\n\ntest_it = test_datagen.flow_from_directory(\n    test_path,\n    class_mode='binary',\n    batch_size=BATCH_SIZE,\n    target_size=(IMAGE_SIZE, IMAGE_SIZE)\n)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:32.603461Z","iopub.execute_input":"2026-10-03T15:21:32.604358Z","iopub.status.idle":"2026-10-03T15:21:45.412256Z","shell.execute_reply.started":"2026-10-03T15:21:32.604326Z","shell.execute_reply":"2026-10-03T15:21:45.411535Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndef plot_exmaple(images, labels):\n    # Assuming images shape is (64, height, width, channels)\n    fig, axes = plt.subplots(8, 8, figsize=(16, 16))\n    axes = axes.flatten()  # Flatten the 8x8 grid into a 1D array for easy iteration\n\n    for i, ax in enumerate(axes):\n        if i < len(images):\n            ax.imshow(images[i])\n            ax.set_title(str(labels[i]))  # Fixed: set_title instead of title\n            # ax.axis(\"off\")  # Turn off axis ticks/labels for clarity\n            ax.axis(\"on\")\n        else:\n            ax.axis(\"off\")  # Hide unused subplots if images < 64\n\n    plt.tight_layout()\n    plt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:45.41326Z","iopub.execute_input":"2026-10-03T15:21:45.413812Z","iopub.status.idle":"2026-10-03T15:21:45.419412Z","shell.execute_reply.started":"2026-10-03T15:21:45.413783Z","shell.execute_reply":"2026-10-03T15:21:45.418441Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_images, train_labels = train_it.__next__()\nprint(train_images.shape, len(train_labels))\nplot_exmaple(train_images, train_labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:45.421096Z","iopub.execute_input":"2026-10-03T15:21:45.421418Z","iopub.status.idle":"2026-10-03T15:21:51.539405Z","shell.execute_reply.started":"2026-10-03T15:21:45.421389Z","shell.execute_reply":"2026-10-03T15:21:51.538374Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"val_images, val_labels = val_it.__next__()\nval_images.shape\nplot_exmaple(val_images, val_labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:51.540673Z","iopub.execute_input":"2026-10-03T15:21:51.541002Z","iopub.status.idle":"2026-10-03T15:21:57.338306Z","shell.execute_reply.started":"2026-10-03T15:21:51.540974Z","shell.execute_reply":"2026-10-03T15:21:57.337395Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def plot_exmaple_test(images, indices):\n    # Assuming images shape is (64, height, width, channels)\n    fig, axes = plt.subplots(8, 8, figsize=(16, 16))\n    axes = axes.flatten()  # Flatten the 8x8 grid into a 1D array for easy iteration\n\n    for i, ax in enumerate(axes):\n        if i < len(images):\n            ax.imshow(images[i])\n            ax.set_title(str(indices[i]))  # ตั้งค่า title ด้วย index จาก list/array indices\n            ax.axis(\"off\")  # Turn off axis ticks/labels for clarity\n        else:\n            ax.axis(\"off\")  # Hide unused subplots if images < 64\n\n    plt.tight_layout()\n    plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:57.339488Z","iopub.execute_input":"2026-10-03T15:21:57.339819Z","iopub.status.idle":"2026-10-03T15:21:57.347122Z","shell.execute_reply.started":"2026-10-03T15:21:57.33979Z","shell.execute_reply":"2026-10-03T15:21:57.346227Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"test_images, test_labels = test_it.__next__()  # หรือ next(test_it)\nprint(f\"{test_images.shape=}, {test_labels.shape}\")\n\n# สร้าง list เลขดัชนีตามจำนวนภาพที่มี\nindices = list(range(len(test_images)))\n\nplot_exmaple(test_images, test_labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:21:57.348254Z","iopub.execute_input":"2026-10-03T15:21:57.34912Z","iopub.status.idle":"2026-10-03T15:22:03.111313Z","shell.execute_reply.started":"2026-10-03T15:21:57.349081Z","shell.execute_reply":"2026-10-03T15:22:03.110291Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# import numpy as np\n\n\n# # 1. สุ่ม index มา 10 ตัวจากชุดข้อมูล\n# sample_indices = np.random.choice(len(test_images), size=10, replace=False)\n# sample_images = test_images[sample_indices]\n\n# # 2. Predict \n# predictions = model_1.predict(sample_images)\n# predicted_classes = np.argmax(predictions, axis=1)  # ดึง class ที่ได้คะแนนสูงสุด\n\n# # 3. แสดงผล 10 ภาพ (ปรับเป็น grid 2x5)\n# fig, axes = plt.subplots(2, 5, figsize=(15, 6))\n# axes = axes.flatten()\n\n# for i, ax in enumerate(axes):\n#     ax.imshow(sample_images[i])\n#     # แสดง Index เดิม และ Class ที่ทายได้\n#     ax.set_title(f\"Idx: {sample_indices[i]}\\nPred: {predicted_classes[i]}\")\n#     ax.axis(\"off\")\n\n# plt.tight_layout()\n# plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:03.112421Z","iopub.execute_input":"2026-10-03T15:22:03.112814Z","iopub.status.idle":"2026-10-03T15:22:03.117551Z","shell.execute_reply.started":"2026-10-03T15:22:03.112789Z","shell.execute_reply":"2026-10-03T15:22:03.116514Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# 2026-09-13\n# Transfer Learning","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:03.118905Z","iopub.execute_input":"2026-10-03T15:22:03.119129Z","iopub.status.idle":"2026-10-03T15:22:03.137433Z","shell.execute_reply.started":"2026-10-03T15:22:03.119109Z","shell.execute_reply":"2026-10-03T15:22:03.13667Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"id(train_it), id(val_it), id(test_it)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:03.13857Z","iopub.execute_input":"2026-10-03T15:22:03.138937Z","iopub.status.idle":"2026-10-03T15:22:03.151093Z","shell.execute_reply.started":"2026-10-03T15:22:03.138907Z","shell.execute_reply":"2026-10-03T15:22:03.15006Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import sys\nfrom matplotlib import pyplot\nfrom keras.utils import to_categorical\nfrom keras.applications.vgg16 import VGG16\nfrom keras.models import Model\nfrom keras.layers import Dense, Flatten, GlobalAveragePooling2D, Dropout # เพิ่ม import ตรงนี้\nfrom keras.optimizers import SGD\nfrom keras.preprocessing.image import ImageDataGenerator\n\n# Hyperparameter => changable\nIMAGE_SIZE = 224\nBATCH_SIE = 64\nEPOCHS = 20\n\n# define cnn model\ndef define_model_vgg16():\n\t# load model\n\tmodel = VGG16(include_top=False, input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3))\n\t\n\t# mark loaded layers as not trainable\n\tfor layer in model.layers:\n\t\tlayer.trainable = False\n\t\t\n\t# --- จุดที่แก้ไข: เปลี่ยน Flatten เป็น GlobalAveragePooling2D ---\n\tx = GlobalAveragePooling2D()(model.layers[-1].output)\n\t\n\t# เพิ่ม Dropout (ปิดการทำงานโหนด 50%) ช่วยลด Overfitting ทำให้คะแนน Test ดีขึ้น\n\tx = Dropout(0.5)(x)\n\t\n\tclass1 = Dense(128, activation='relu', kernel_initializer='he_uniform')(x)\n\toutput = Dense(1, activation='sigmoid')(class1)\n\t\n\t# define new model\n\tmodel = Model(inputs=model.inputs, outputs=output)\n\t\n\t# compile model\n\topt = SGD(learning_rate=0.001, momentum=0.9) # หมายเหตุ: Keras เวอร์ชั่นใหม่ แนะนำให้ใช้ learning_rate แทน lr\n\tmodel.compile(optimizer=opt, loss='binary_crossentropy', metrics=['accuracy'])\n\t\n\treturn model","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:03.152103Z","iopub.execute_input":"2026-10-03T15:22:03.152434Z","iopub.status.idle":"2026-10-03T15:22:03.160668Z","shell.execute_reply.started":"2026-10-03T15:22:03.152411Z","shell.execute_reply":"2026-10-03T15:22:03.159772Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Load model\nmodel_vgg16_top = VGG16(include_top=True,\n                        input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3)\n                       )\nmodel_vgg16_top.summary()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:03.161837Z","iopub.execute_input":"2026-10-03T15:22:03.162098Z","iopub.status.idle":"2026-10-03T15:22:07.60141Z","shell.execute_reply.started":"2026-10-03T15:22:03.162059Z","shell.execute_reply":"2026-10-03T15:22:07.600417Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Load model\nmodel_16_feature_extractor = VGG16(include_top=False,\n                                   input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3)\n                                  )\nmodel_16_feature_extractor.summary() #feature extractor as a backbone","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:07.603816Z","iopub.execute_input":"2026-10-03T15:22:07.604161Z","iopub.status.idle":"2026-10-03T15:22:08.322553Z","shell.execute_reply.started":"2026-10-03T15:22:07.604138Z","shell.execute_reply":"2026-10-03T15:22:08.321858Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for index, layer in enumerate(model_16_feature_extractor.layers):\n\t\tprint(index, \n              layer.name,\n              layer.trainable,\n              layer.output.shape\n        )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:08.323707Z","iopub.execute_input":"2026-10-03T15:22:08.324054Z","iopub.status.idle":"2026-10-03T15:22:08.330337Z","shell.execute_reply.started":"2026-10-03T15:22:08.324012Z","shell.execute_reply":"2026-10-03T15:22:08.329028Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for index, layer in enumerate(model_16_feature_extractor.layers):\n    if index <= 14:\n        layer.trainable = False\n\nfor index, layer in enumerate(model_16_feature_extractor.layers):\n    print(index, layer.name, layer.trainable)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:08.331467Z","iopub.execute_input":"2026-10-03T15:22:08.331775Z","iopub.status.idle":"2026-10-03T15:22:08.345206Z","shell.execute_reply.started":"2026-10-03T15:22:08.331742Z","shell.execute_reply":"2026-10-03T15:22:08.344423Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model_16_feature_extractor.layers[-1].output","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:22:08.346406Z","iopub.execute_input":"2026-10-03T15:22:08.346793Z","iopub.status.idle":"2026-10-03T15:22:08.359413Z","shell.execute_reply.started":"2026-10-03T15:22:08.346762Z","shell.execute_reply":"2026-10-03T15:22:08.358643Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from keras.layers import GlobalAveragePooling2D, Dense\n\n# 1. แปลงรูปทรงของ output เดิมด้วย GAP (เราตั้งชื่อตัวแปรใหม่ว่า pooled_output มาใช้แทน flat1)\npooled_output = GlobalAveragePooling2D()(model_16_feature_extractor.layers[-1].output)\n\n# 2. สร้าง layer1 โดยรับค่ามาจาก pooled_output ด้านบน\nlayer1 = Dense(256, activation='relu', kernel_initializer='he_uniform')(pooled_output)\nprint(layer1.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.001374Z","iopub.execute_input":"2026-10-03T15:24:14.001914Z","iopub.status.idle":"2026-10-03T15:24:14.025918Z","shell.execute_reply.started":"2026-10-03T15:24:14.001883Z","shell.execute_reply":"2026-10-03T15:24:14.024963Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"layer2 = Dense(128, activation='relu', kernel_initializer='he_uniform')(layer1)\nprint(layer2.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.027412Z","iopub.execute_input":"2026-10-03T15:24:14.027963Z","iopub.status.idle":"2026-10-03T15:24:14.043817Z","shell.execute_reply.started":"2026-10-03T15:24:14.027937Z","shell.execute_reply":"2026-10-03T15:24:14.043056Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Use functional API\nflat1 = Flatten()(model_16_feature_extractor.layers[-1].output)\nprint(flat1.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.044813Z","iopub.execute_input":"2026-10-03T15:24:14.045061Z","iopub.status.idle":"2026-10-03T15:24:14.054168Z","shell.execute_reply.started":"2026-10-03T15:24:14.04504Z","shell.execute_reply":"2026-10-03T15:24:14.053354Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"output1 = Dense(2, activation='softmax')(layer2)\nprint(f\"{output1.shape=}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.055241Z","iopub.execute_input":"2026-10-03T15:24:14.05559Z","iopub.status.idle":"2026-10-03T15:24:14.072338Z","shell.execute_reply.started":"2026-10-03T15:24:14.055568Z","shell.execute_reply":"2026-10-03T15:24:14.071273Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"output2 = Dense(1, activation='sigmoid')(layer2)\nprint(output2.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.074784Z","iopub.execute_input":"2026-10-03T15:24:14.075198Z","iopub.status.idle":"2026-10-03T15:24:14.092072Z","shell.execute_reply.started":"2026-10-03T15:24:14.075172Z","shell.execute_reply":"2026-10-03T15:24:14.09106Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# define new model to be trained\nmodel_vgg16 = Model(inputs=model_16_feature_extractor.inputs, outputs=output2)\n# compile model\nopt = SGD(lr=0.001, momentum=0.9)\n\nmodel_vgg16.compile(optimizer=opt,\n                    loss='binary_crossentropy',\n                    metrics=['accuracy']\n                   )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.093146Z","iopub.execute_input":"2026-10-03T15:24:14.093856Z","iopub.status.idle":"2026-10-03T15:24:14.108294Z","shell.execute_reply.started":"2026-10-03T15:24:14.093822Z","shell.execute_reply":"2026-10-03T15:24:14.10758Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"for index, layer in enumerate(model_vgg16.layers):\n    print(index, layer.name, layer.trainable)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.109179Z","iopub.execute_input":"2026-10-03T15:24:14.109509Z","iopub.status.idle":"2026-10-03T15:24:14.114882Z","shell.execute_reply.started":"2026-10-03T15:24:14.10948Z","shell.execute_reply":"2026-10-03T15:24:14.113792Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"print(model_vgg16.summary())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.11593Z","iopub.execute_input":"2026-10-03T15:24:14.116753Z","iopub.status.idle":"2026-10-03T15:24:14.162743Z","shell.execute_reply.started":"2026-10-03T15:24:14.116675Z","shell.execute_reply":"2026-10-03T15:24:14.161675Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"id(train_it), id(val_it)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.1637Z","iopub.execute_input":"2026-10-03T15:24:14.163996Z","iopub.status.idle":"2026-10-03T15:24:14.169926Z","shell.execute_reply.started":"2026-10-03T15:24:14.163975Z","shell.execute_reply":"2026-10-03T15:24:14.168838Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# del ตัวแปร #ลบตัวแปรออกจากหน่วยความจำ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.170969Z","iopub.execute_input":"2026-10-03T15:24:14.17124Z","iopub.status.idle":"2026-10-03T15:24:14.178786Z","shell.execute_reply.started":"2026-10-03T15:24:14.171219Z","shell.execute_reply":"2026-10-03T15:24:14.177875Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# model_vgg16 = define_model_vgg16()\n# print(model_vgg16)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.17987Z","iopub.execute_input":"2026-10-03T15:24:14.180646Z","iopub.status.idle":"2026-10-03T15:24:14.194237Z","shell.execute_reply.started":"2026-10-03T15:24:14.180622Z","shell.execute_reply":"2026-10-03T15:24:14.193166Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# fit model\nimport tensorflow as tf\nfrom keras.callbacks import ReduceLROnPlateau, EarlyStopping\nfrom keras.optimizers import SGD\n\n# 1. กำหนด Callbacks ของเรา (เพิ่มเข้ามาใหม่)\nrlr = ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=2, verbose=1, min_lr=1e-6)\nes = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True, verbose=1)\n\n# 2. Verify that TensorFlow can communicate with the GPU hardware\nphysical_devices = tf.config.list_physical_devices('GPU')\nprint(f\"Hardware status: {len(physical_devices)} GPU(s) detected.\")\n\nif len(physical_devices) > 0:\n    # Optional but highly recommended: Prevent TF from pre-allocating 100% of VRAM\n    try:\n        tf.config.experimental.set_memory_growth(physical_devices[0], True)\n    except Exception as e:\n        print(f\"Memory growth configuration failed: {e}\")\n\n    # Explicitly bind the execution to the primary GPU\n    with tf.device('/GPU:0'):\n        \n        print(f\"Model ID: {id(model_vgg16)}\")\n        model_vgg16.reset_states() \n        \n        # ==========================================\n        # Phase 1: Warm-up (เทรนเฉพาะส่วน Head ที่เพิ่งเปลี่ยนเป็น GAP)\n        # ==========================================\n        print(\"\\n--- Phase 1: Training Custom Head ---\")\n        model_vgg16_history = model_vgg16.fit(\n            train_it, \n            steps_per_epoch=len(train_it),\n            validation_data=val_it,\n            validation_steps=len(val_it),\n            epochs=10, # รันรอบแรกสัก 10 รอบให้พอนิ่ง\n            verbose=1,\n            # รวม Plot ของอาจารย์ เข้ากับ Callbacks ดันคะแนนของเรา\n            callbacks=[PlotLossesKeras(), rlr, es] \n        )\n\n        # ==========================================\n        # Phase 2: Fine-Tuning (ปลดล็อก VGG16 เลเยอร์ลึกสุด)\n        # ==========================================\n        print(\"\\n--- Phase 2: Fine-Tuning VGG16 Block 5 ---\")\n        \n        # ปลดล็อก VGG16 \n        model_vgg16.trainable = True\n        \n        # ล็อกชั้นอื่นไว้ ปลดล็อกเฉพาะชั้นที่ชื่อขึ้นต้นด้วย 'block5'\n        for layer in model_vgg16.layers:\n            if layer.name.startswith('block5'):\n                layer.trainable = True\n            else:\n                layer.trainable = False\n                \n        # Compile ใหม่ **ต้องใช้ Learning Rate ที่ต่ำมากๆ (1e-5)**\n        opt_fine = SGD(learning_rate=1e-5, momentum=0.9)\n        model_vgg16.compile(optimizer=opt_fine, loss='binary_crossentropy', metrics=['accuracy'])\n        \n        # สั่งเทรนต่อเพื่อรีดคะแนนให้ทะลุ 0.97\n        history_fine = model_vgg16.fit(\n            train_it,\n            steps_per_epoch=len(train_it),\n            validation_data=val_it,\n            validation_steps=len(val_it),\n            epochs=15, \n            verbose=1,\n            callbacks=[PlotLossesKeras(), rlr, es]\n        )\n\nelse:\n    print(\"WARNING: No GPU detected. Execution will proceed on the CPU, which will be significantly slower.\")\n    # Fallback to standard execution (ของเดิมอาจารย์)\n    model_vgg16_history = model_vgg16.fit(\n        train_it, \n        steps_per_epoch=len(train_it),\n        validation_data=val_it,\n        validation_steps=len(val_it),\n        epochs=2, \n        verbose=1,\n        callbacks=[PlotLossesKeras()]\n    )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T15:24:14.249798Z","iopub.execute_input":"2026-10-03T15:24:14.250124Z","iopub.status.idle":"2026-10-03T16:53:14.847392Z","shell.execute_reply.started":"2026-10-03T15:24:14.250094Z","shell.execute_reply":"2026-10-03T16:53:14.846387Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# learning curves\nsummarize_diagnostics(model_vgg16_history)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:53:14.849896Z","iopub.execute_input":"2026-10-03T16:53:14.85017Z","iopub.status.idle":"2026-10-03T16:53:15.228034Z","shell.execute_reply.started":"2026-10-03T16:53:14.850147Z","shell.execute_reply":"2026-10-03T16:53:15.226943Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# evaluate model\n_, acc = model_vgg16.evaluate(val_it, steps=len(val_it), verbose=1)\nprint('Val: > %.3f' % (acc * 100.0))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:53:15.229437Z","iopub.execute_input":"2026-10-03T16:53:15.230469Z","iopub.status.idle":"2026-10-03T16:53:31.215146Z","shell.execute_reply.started":"2026-10-03T16:53:15.230433Z","shell.execute_reply":"2026-10-03T16:53:31.21407Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# evaluate model for the test set\n_, test_acc = model_vgg16.evaluate(test_it, steps=len(test_it), verbose=1)\nprint('Test: > %.3f' % (test_acc * 100.0))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:53:31.216309Z","iopub.execute_input":"2026-10-03T16:53:31.216889Z","iopub.status.idle":"2026-10-03T16:54:29.129343Z","shell.execute_reply.started":"2026-10-03T16:53:31.216862Z","shell.execute_reply":"2026-10-03T16:54:29.128592Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import numpy as np\nfrom sklearn.metrics import classification_report\n\n# 1. Reset the generator state to ensure alignment from index 0\ntest_it.reset()\n\n# 2. Extract the ordered ground truth labels from the generator\ny_true = test_it.classes\nprint(y_true.shape)\n\n# 3. Execute the forward pass over the entire test dataset\n# Calculate exact steps to prevent infinite looping or partial evaluation\nsteps = np.ceil(test_it.samples / test_it.batch_size)\nprobabilities = model_vgg16.predict(test_it, steps=steps, verbose=1).squeeze()\nprint(probabilities.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:54:29.131968Z","iopub.execute_input":"2026-10-03T16:54:29.132206Z","iopub.status.idle":"2026-10-03T16:55:10.715616Z","shell.execute_reply.started":"2026-10-03T16:54:29.132185Z","shell.execute_reply":"2026-10-03T16:55:10.714601Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"probabilities[:20]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:55:10.716861Z","iopub.execute_input":"2026-10-03T16:55:10.717213Z","iopub.status.idle":"2026-10-03T16:55:10.724379Z","shell.execute_reply.started":"2026-10-03T16:55:10.717189Z","shell.execute_reply":"2026-10-03T16:55:10.723215Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"(probabilities[0] > 0.5).astype(int)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:55:10.725502Z","iopub.execute_input":"2026-10-03T16:55:10.725855Z","iopub.status.idle":"2026-10-03T16:55:10.73952Z","shell.execute_reply.started":"2026-10-03T16:55:10.725825Z","shell.execute_reply":"2026-10-03T16:55:10.738553Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# 4. Discretize continuous logistic probabilities into binary class predictions\ny_pred = (probabilities > 0.5).astype(int)\nprint(y_pred.shape)\n\n# 5. Extract dynamic class labels mapping (e.g., {'cats': 0, 'dogs': 1})\n# This ensures the report correctly maps index 0 and 1 to your string labels\ntarget_names = list(test_it.class_indices.keys())\n\n# 6. Compute and output the classification report\nreport = classification_report(y_true, y_pred, target_names=target_names)\nprint(report)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:55:10.740513Z","iopub.execute_input":"2026-10-03T16:55:10.740821Z","iopub.status.idle":"2026-10-03T16:55:10.769726Z","shell.execute_reply.started":"2026-10-03T16:55:10.740791Z","shell.execute_reply":"2026-10-03T16:55:10.769039Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"y_true[:10], y_pred[:10]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:55:10.770575Z","iopub.execute_input":"2026-10-03T16:55:10.770851Z","iopub.status.idle":"2026-10-03T16:55:10.776822Z","shell.execute_reply.started":"2026-10-03T16:55:10.770829Z","shell.execute_reply":"2026-10-03T16:55:10.775776Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Predict\nimport numpy as np\nimport matplotlib.pyplot as plt\n\n# 1. Aggregate sufficient samples from the iterator\nimages_list = []\nlabels_list = []\n\n# Fetch 2 batches (2 * 64 = 128 samples) to ensure we have pool > 100\nfor _ in range(4):\n    batch_images, batch_labels = next(test_it) # หากโค้ดก่อนหน้าใช้ตัวแปรชื่อ val_it ให้เปลี่ยนเป็น next(val_it) นะครับ\n    images_list.append(batch_images)\n    labels_list.append(batch_labels)\n\n# Concatenate into a single tensor\nall_images = np.concatenate(images_list, axis=0)\nall_labels = np.concatenate(labels_list, axis=0)\n\n# 2. Stochastically sample 100 unique indices\nnp.random.seed(42) # For reproducibility\nrandom_indices = np.random.choice(all_images.shape[0], size=100, replace=False)\n\nsample_images = all_images[random_indices]\nsample_labels = all_labels[random_indices]\n\n# 3. Execute the forward pass\n# Squeeze is used to convert shape (100, 1) to (100,)\nprobabilities = model_vgg16.predict(sample_images).squeeze()\n\n# 4. Thresholding (Assuming standard mapping: 0 = Cat, 1 = Dog)\npredictions = (probabilities > 0.5).astype(int)\nclass_mapping = {0: 'Cat', 1: 'Dog'}\n\nprint(\"Done.\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:57:59.244798Z","iopub.execute_input":"2026-10-03T16:57:59.245152Z","iopub.status.idle":"2026-10-03T16:58:01.60008Z","shell.execute_reply.started":"2026-10-03T16:57:59.245129Z","shell.execute_reply":"2026-10-03T16:58:01.599355Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Initialize a 10x10 matrix visualization\nfig, axes = plt.subplots(10, 10, figsize=(20, 22))\naxes = axes.flatten()\n\nfor i in range(100):\n    ax = axes[i]\n    \n    # Isolate image tensor. \n    # NOTE: If your VGG16 preprocessor applies mean-centering (caffe style) \n    # instead of standard [0,1] scaling, you must reverse it here for correct plt.imshow() rendering.\n    # We assume standard [0, 1] or [0, 255] spatial tensors below.\n    img = sample_images[i]\n    if img.max() > 1.0 or img.min() < 0:\n        # Simple normalization for display purposes\n        img = (img - img.min()) / (img.max() - img.min())\n        \n    ax.imshow(img)\n    ax.axis('off')\n    \n    # Extract structural labels\n    gt_class = class_mapping[int(sample_labels[i])]\n    pred_class = class_mapping[predictions[i]]\n    prob = probabilities[i]\n    \n    # Apply conditional color coding (Green for True Positives/Negatives, Red for False Positives/Negatives)\n    text_color = 'darkgreen' if gt_class == pred_class else 'darkred'\n    \n    title_str = f\"GT: {gt_class}\\nPred: {pred_class}\\nProb: {prob:.3f}\"\n    ax.set_title(title_str, color=text_color, fontsize=9, pad=4)\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:58:07.075864Z","iopub.execute_input":"2026-10-03T16:58:07.076325Z","iopub.status.idle":"2026-10-03T16:58:14.957263Z","shell.execute_reply.started":"2026-10-03T16:58:07.076297Z","shell.execute_reply":"2026-10-03T16:58:14.955788Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"stop","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T16:55:11.024121Z","iopub.status.idle":"2026-10-03T16:55:11.024401Z","shell.execute_reply.started":"2026-10-03T16:55:11.024266Z","shell.execute_reply":"2026-10-03T16:55:11.02428Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# InceptionV3","metadata":{}},{"cell_type":"code","source":"from keras.applications.inception_v3 import InceptionV3\nfrom keras.layers import GlobalAveragePooling2D, Dense, Dropout\nfrom keras.models import Model\nfrom keras.optimizers import SGD\n\ndef define_model_inceptionv3():\n    # 1. เปลี่ยนขนาดภาพให้รับค่าตัวแปร IMAGE_SIZE (224x224) ตาม Generator ของเรา\n    model = InceptionV3(include_top=False, input_shape=(IMAGE_SIZE, IMAGE_SIZE, 3))\n    \n    # ล็อก weights ของ InceptionV3 ไว้ก่อน\n    for layer in model.layers:\n        layer.trainable = False\n        \n    # 2. เปลี่ยน Flatten เป็น GlobalAveragePooling2D เพื่อลดพารามิเตอร์และกัน Overfit\n    x = GlobalAveragePooling2D()(model.layers[-1].output)\n    \n    # เพิ่ม Dropout ปิดโหนด 50% ตอนเทรน ช่วยให้โมเดลไม่จำข้อสอบ\n    x = Dropout(0.5)(x)\n    \n    # ยุบ Dense layer ให้เหลือชั้นเดียวก็เพียงพอสำหรับแยกหมา-แมวแล้วครับ\n    class1 = Dense(256, activation='relu', kernel_initializer='he_uniform')(x)\n    \n    # 3. แก้ Output ให้เป็น 1 โหนด และใช้ sigmoid เพื่อแก้ Error\n    output = Dense(1, activation='sigmoid')(class1)\n    \n    # define new model\n    model = Model(inputs=model.inputs, outputs=output)\n    \n    # compile model (เปลี่ยน lr เป็น learning_rate ป้องกัน Keras แจ้งเตือน)\n    opt = SGD(learning_rate=0.001, momentum=0.9)\n    model.compile(optimizer=opt, loss='binary_crossentropy', metrics=['accuracy'])\n    \n    return model","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:05:54.00823Z","iopub.execute_input":"2026-10-03T17:05:54.0088Z","iopub.status.idle":"2026-10-03T17:05:54.016002Z","shell.execute_reply.started":"2026-10-03T17:05:54.008765Z","shell.execute_reply":"2026-10-03T17:05:54.015051Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model_inceptionv3 = define_model_inceptionv3()\nprint(id(model_inceptionv3))\n# model_inceptionv3.summary()","metadata":{"execution":{"iopub.status.busy":"2026-10-03T17:06:10.612856Z","iopub.execute_input":"2026-10-03T17:06:10.613534Z","iopub.status.idle":"2026-10-03T17:06:12.708845Z","shell.execute_reply.started":"2026-10-03T17:06:10.613503Z","shell.execute_reply":"2026-10-03T17:06:12.708021Z"},"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import numpy as np\nfrom tensorflow.keras import backend as K\n\ntrainable_params = int(np.sum([K.count_params(w) for w in model_inceptionv3.trainable_weights]))\nprint(trainable_params)\n","metadata":{"trusted":true,"scrolled":true,"execution":{"iopub.status.busy":"2026-10-03T17:06:17.779998Z","iopub.execute_input":"2026-10-03T17:06:17.780482Z","iopub.status.idle":"2026-10-03T17:06:17.78711Z","shell.execute_reply.started":"2026-10-03T17:06:17.780452Z","shell.execute_reply":"2026-10-03T17:06:17.78614Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from tensorflow.keras import backend as K\n\nfor w in model_inceptionv3.trainable_weights:\n    print(w.name, K.count_params(w))\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:06:20.639875Z","iopub.execute_input":"2026-10-03T17:06:20.640339Z","iopub.status.idle":"2026-10-03T17:06:20.646456Z","shell.execute_reply.started":"2026-10-03T17:06:20.64031Z","shell.execute_reply":"2026-10-03T17:06:20.645501Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# fit model\nmodel_inceptionv3.reset_states()\nhistory_inceptionv3 = model_inceptionv3.fit(\n    train_it, \n    steps_per_epoch=len(train_it),\n    validation_data=val_it,\n    validation_steps=len(val_it),\n    epochs=3, verbose=1,\n    callbacks = [PlotLossesKeras()]\n    )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:06:25.516769Z","iopub.execute_input":"2026-10-03T17:06:25.517078Z","iopub.status.idle":"2026-10-03T17:16:41.562983Z","shell.execute_reply.started":"2026-10-03T17:06:25.517056Z","shell.execute_reply":"2026-10-03T17:16:41.562011Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndef plot_images_with_labels(images, labels, train_log, set_label, save=False):\n    \"\"\"\n    Plots a grid of images with their corresponding labels and saves the figure.\n    \n    Args:\n    - images (list or np.array): List of images to be displayed.\n    - labels (np.array): Array of one-hot encoded labels.\n    - train_log (str): The directory or log name to save the plot.\n    - set_label (str): The specific set label (e.g., \"train\", \"validation\", etc.).\n    - config (dict): Configuration dictionary containing image dimensions.\n    \n    Returns:\n    - None: Displays and saves the plot.\n    \"\"\"\n    # Convert one-hot encoded labels to class indices\n    class_indices = labels\n\n    # Create subplots\n    fig, axes = plt.subplots(3, 4, figsize=(14, 10), dpi=120)\n    axes = axes.flatten()\n\n    for img, label_idx, ax in zip(images, class_indices, axes):\n        # Undo any preprocessing (like rescaling for pre-trained models)\n        if np.max(img) == 1:\n            img = img * 255\n\n        # Display the image\n        img_shape = (224, 224, 3)\n        ax.imshow(img.astype(\"uint8\"))\n        # ax.imshow(img)\n\n        # Set the title to the corresponding class label\n        ax.set_title(f\"Label: {label_idx}\", fontsize=18)\n        ax.axis('on')\n\n    # Set the overall title for the figure\n    plot_title = f\"{train_log}-{set_label}\"\n    plt.suptitle(plot_title, fontsize=18)\n\n    # Show the plot\n    plt.show()\n\n    # Save the figure\n    if save:\n        plot_file = os.path.join(train_log, f\"{plot_title}.png\")\n        fig.savefig(plot_file, dpi=120)\n        print(f'Saved {plot_file}')\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:17:06.204643Z","iopub.execute_input":"2026-10-03T17:17:06.20517Z","iopub.status.idle":"2026-10-03T17:17:06.212768Z","shell.execute_reply.started":"2026-10-03T17:17:06.205141Z","shell.execute_reply":"2026-10-03T17:17:06.211767Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# plot_images_with_labels(images, labels, \"Dog-Cat\", \"Training\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:17:11.338074Z","iopub.execute_input":"2026-10-03T17:17:11.338825Z","iopub.status.idle":"2026-10-03T17:17:11.342526Z","shell.execute_reply.started":"2026-10-03T17:17:11.338786Z","shell.execute_reply":"2026-10-03T17:17:11.341566Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# for index, layer in enumerate(model.layers):\n#     print(index, layer.output.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:17:12.731222Z","iopub.execute_input":"2026-10-03T17:17:12.731802Z","iopub.status.idle":"2026-10-03T17:17:12.735957Z","shell.execute_reply.started":"2026-10-03T17:17:12.731772Z","shell.execute_reply":"2026-10-03T17:17:12.735055Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# # mark loaded layers as not trainable\n# for index, layer in enumerate(model.layers[:15]):\n#     print(index, layer.name, layer.output_shape)\n#     layer.trainable = False\n    \n# for index, layer in enumerate(model.layers[15:]):\n#     print(index, layer.name, layer.output_shape)\n#     layer.trainable = True","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:17:13.708332Z","iopub.execute_input":"2026-10-03T17:17:13.709038Z","iopub.status.idle":"2026-10-03T17:17:13.712926Z","shell.execute_reply.started":"2026-10-03T17:17:13.709006Z","shell.execute_reply":"2026-10-03T17:17:13.711928Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# evaluate model\n_, acc = model_inceptionv3.evaluate(test_it, steps=len(test_it), verbose=1)\nprint('> %.3f' % (acc * 100.0))\n# learning curves\n# summarize_diagnostics(history)\n\n# best_model = 'vgg16_model.hdf5'\n# my_model.save(best_model)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:17:17.384865Z","iopub.execute_input":"2026-10-03T17:17:17.385623Z","iopub.status.idle":"2026-10-03T17:17:43.952873Z","shell.execute_reply.started":"2026-10-03T17:17:17.385591Z","shell.execute_reply":"2026-10-03T17:17:43.951952Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# แนะนำให้ใช้ฟังก์ชัน next() ซึ่งเป็นมาตรฐานของ Python แทนการเรียก .__next__() ตรงๆ\nbatch_images, batch_labels = next(test_it)\n\n# ตอนนี้ batch_images จะเป็น Numpy array ของรูปภาพแล้ว สามารถเรียกดู shape ได้เลย\nprint(\"Image shape:\", batch_images.shape)\nprint(\"Label shape:\", batch_labels.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:19:17.750031Z","iopub.execute_input":"2026-10-03T17:19:17.751056Z","iopub.status.idle":"2026-10-03T17:19:17.856287Z","shell.execute_reply.started":"2026-10-03T17:19:17.751022Z","shell.execute_reply":"2026-10-03T17:19:17.855263Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# 5 first img: 1, 0, 1, 1, 0\nimages_to_predict = batch_images[:5]\nimages_to_predict.shape","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:21:06.191452Z","iopub.execute_input":"2026-10-03T17:21:06.191932Z","iopub.status.idle":"2026-10-03T17:21:06.197883Z","shell.execute_reply.started":"2026-10-03T17:21:06.191901Z","shell.execute_reply":"2026-10-03T17:21:06.196978Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"probs = model_inceptionv3.predict(images_to_predict, verbose=1)\nprobs","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T17:21:07.918478Z","iopub.execute_input":"2026-10-03T17:21:07.919281Z","iopub.status.idle":"2026-10-03T17:21:10.260116Z","shell.execute_reply.started":"2026-10-03T17:21:07.919248Z","shell.execute_reply":"2026-10-03T17:21:10.259257Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"Make Prediction","metadata":{}},{"cell_type":"code","source":"!pip install livelossplot","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T21:04:04.253458Z","iopub.execute_input":"2026-10-03T21:04:04.253942Z","iopub.status.idle":"2026-10-03T21:04:14.432764Z","shell.execute_reply.started":"2026-10-03T21:04:04.253908Z","shell.execute_reply":"2026-10-03T21:04:14.431801Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# =============================================================================\n# โมเดลที่ 3: EfficientNetB0 (ตัวทำคะแนน 0.97+)\n# =============================================================================\nimport tensorflow as tf\nfrom keras.applications import EfficientNetB0\nfrom keras.layers import GlobalAveragePooling2D, Dense, Dropout, Input\nfrom keras.models import Model\nfrom keras.optimizers import Adam\nfrom keras.callbacks import ReduceLROnPlateau, EarlyStopping\nfrom livelossplot import PlotLossesKeras\nfrom sklearn.metrics import classification_report\nimport numpy as np\n\n# ป้องกัน Error ตัวแปรหาย\nIMAGE_SIZE = 224\n\n# ---------------------------------------------------------\n# ส่วนที่ 1: ฟังก์ชันสร้างโมเดล\n# ---------------------------------------------------------\ndef define_model_efficientnet():\n    inputs = Input(shape=(IMAGE_SIZE, IMAGE_SIZE, 3))\n    \n    # แก้ปัญหาภาพมืด: ถ่างสเกลพิกเซลจาก 0-1 กลับเป็น 0-255 \n    x = tf.keras.layers.Rescaling(scale=255.0)(inputs)\n    \n    # โหลด Base Model\n    base_model = EfficientNetB0(include_top=False, weights='imagenet', input_tensor=x)\n    \n    # ล็อก Weights ไว้ก่อน\n    for layer in base_model.layers:\n        layer.trainable = False\n        \n    # ต่อ Head \n    x = GlobalAveragePooling2D()(base_model.output)\n    x = Dropout(0.5)(x)\n    class1 = Dense(128, activation='relu', kernel_initializer='he_uniform')(x)\n    output = Dense(1, activation='sigmoid')(class1)\n    \n    model = Model(inputs=inputs, outputs=output)\n    model.compile(optimizer=Adam(learning_rate=0.001), loss='binary_crossentropy', metrics=['accuracy'])\n    \n    return model\n\n# ---------------------------------------------------------\n# ส่วนที่ 2: ตั้งค่า Callbacks และรันโมเดล\n# ---------------------------------------------------------\nrlr = ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=2, verbose=1, min_lr=1e-6)\nes = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True, verbose=1)\n\nmodel_eff = define_model_efficientnet()\n\n# เช็ค GPU แบบของอาจารย์\nphysical_devices = tf.config.list_physical_devices('GPU')\nif len(physical_devices) > 0:\n    with tf.device('/GPU:0'):\n        \n        print(\"\\n=== Phase 1: Training Custom Head (5 Epochs) ===\")\n        model_eff.fit(\n            train_it, \n            steps_per_epoch=len(train_it),\n            validation_data=val_it,\n            validation_steps=len(val_it),\n            epochs=5,\n            verbose=1,\n            callbacks=[PlotLossesKeras(), rlr, es]\n        )\n\n        print(\"\\n=== Phase 2: Fine-Tuning (10 Epochs) ===\")\n        model_eff.trainable = True\n\n        # ล็อกเลเยอร์ 100 ชั้นแรกไว้ ปลดล็อกแค่ครึ่งหลังให้โฟกัสหมาแมว\n        for layer in model_eff.layers[:100]:\n            layer.trainable = False\n\n        # ใช้ Learning Rate ต่ำมากๆ (1e-5) สำหรับ Fine-Tuning\n        model_eff.compile(optimizer=Adam(learning_rate=1e-5), loss='binary_crossentropy', metrics=['accuracy'])\n\n        model_eff.fit(\n            train_it, \n            steps_per_epoch=len(train_it),\n            validation_data=val_it,\n            validation_steps=len(val_it),\n            epochs=10, \n            verbose=1,\n            callbacks=[PlotLossesKeras(), rlr, es]\n        )\nelse:\n    print(\"Warning: No GPU detected. Please turn on GPU in Kaggle.\")\n\n# ---------------------------------------------------------\n# ส่วนที่ 3: สรุปผล Classification Report ส่งอาจารย์\n# ---------------------------------------------------------\nprint(\"\\n=== Generating Classification Report ===\")\nval_it.reset() # สำคัญมาก: รีเซ็ตข้อมูลไม่ให้ลำดับเพี้ยน\nprobs = model_eff.predict(val_it, steps=len(val_it), verbose=1)\npreds = (probs > 0.5).astype(int).reshape(-1)\ny_true = val_it.classes\n\nprint(\"\\n[ Classification Report: EfficientNetB0 ]\")\nprint(classification_report(y_true, preds, target_names=['Cat', 'Dog'], digits=4))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-10-03T21:12:07.633823Z","iopub.execute_input":"2026-10-03T21:12:07.634774Z","iopub.status.idle":"2026-10-03T22:03:33.750051Z","shell.execute_reply.started":"2026-10-03T21:12:07.634702Z","shell.execute_reply":"2026-10-03T22:03:33.748943Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}